Viviannru

Создаем видео, где мы танцуем на парковке.

2026-09-22T11:27:33.071Z

Нужен аккаунт в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
*Подписка PRO и выше.
Нейросети: Grok Imagine 2.0 или Gpt Image 2 для фото, Seedance 2.5 для видео.
Нужно: 3 фото героев ролика, + видео-исходник (ссылка)

1.Создаем карты персонажей.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Grok Imagine 2.0 или Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

ПРОМПТ ДЛЯ КАРТЫ ПЕРСОНАЖА:

Create a photorealistic character reference sheet from the uploaded person photos: three vertical panels, clean white or light-gray background, clear vertical dividers.

RIGHT: Shoulder-cropped frontal portrait, direct gaze, neutral expression, soft even light. Preserve identity: facial proportions, eyes, nose, lips, brows, ears, skin tone and texture, hair color and shape, moles and freckles. Highly detailed, natural realism.

LEFT: Front view, framed from the neck/shoulder line down to the feet; head and face outside the crop. Standing upright, arms at sides, feet shoulder-width apart, wearing the person's everyday clothing. Preserve build, proportions, and outfit details.

MIDDLE: Full-body BACK view, including the head and hair from behind. Same person, build, proportions, and outfit.

Keep identity, clothing, lighting, and photographic style consistent across panels. Match body scale in the left and middle panels. High resolution, realistic anatomy and fabric.

*С остальными фотками тоже самое.

2.Создаем видео.

Для начала скачиваем видео-исходник: ССЫЛКА

- Дальше переходим в Раздел Видео
- Выбираем Seedance 2.5
- Режим Omni Reference (или EDIT)
- Качество 720p (480р, чтобы сэкономить, а если мажоры, то 1080р)
- Соотношение сторон 9:16 (если EDIT, то "авто")
- Длительность 20 сек. (если EDIT, то "авто").
- Загружаем карты персонажей + исходник видео.
- После вставляем промпт и вручную удаляем все "@image1", "@image2", "@image3", "@video1" и заново их прописываем, когда будете вводить "@" там можно будет выбрать нужный файл, так они привяжутся к промпту *Если делаете в Syntx.ai, то ничего делать не нужно, они автоматически привяжутся:

ПРОМПТ ДЛЯ ВИДЕО (РЕЖДИМ - OMNI REFERENCE):

[Generation Task — New Video, Not Source Editing] Generate a NEW photorealistic video using @video1 as a reference for scene layout, action choreography, emotional performances, camera movement, and overall pacing. Do NOT edit, repaint, or face-swap the original video frames. Do NOT place reference faces onto the original actors' heads or bodies. Build three complete, coherent people from the photographic references and render them naturally within a newly generated scene. Target duration: approximately 20 seconds. Aspect ratio: 9:00 — use the standard vertical 9:16 presentation. [Reference Roles and Priority] @image1 = complete appearance of PERSON_A. @image2 = complete appearance of PERSON_B. @image3 = complete appearance of these PERSON_C. @video1 = reference for: - Location and spatial relationships. - Sequence of events. - Human actions and gestures. - Facial expressions and emotional changes. - Camera positions, handheld movements, turns, and zooms. - Approximate action timing and overall visual rhythm. - Evening lighting and atmosphere. @video1 is NOT a reference for: - Character identities. - Head shapes, facial proportions, or hairstyles. - Body types or gender presentation. - Clothing, hats, or wearable accessories. - Captions, lettering, emojis, logos, or other graphics. If the source actors' anatomy conflicts with the photographic references, prioritize the photographic references. Adapt the choreography rather than distorting the reference person to fit the source actor. [Complete Character Construction] Construct each person from their assigned photograph as a unified head and body, not a face mask. Preserve the reference person's: - Overall head shape and cranial proportions. - Face width and length, forehead, cheekbones, jawline, and chin. - Eye shape and spacing, nose shape, mouth shape, and ears. - Apparent age, gender presentation, and skin tone. - Hairline, hairstyle, hair color, and facial hair if present. - Visible neck proportions, shoulder width, and physique. - Clothing, garment fit, colors, materials, and layers. - Headwear and wearable accessories, where present. - Footwear, where visible. Do not inherit the original actors' skull shapes, jawlines, body silhouettes, or characteristic facial features. Maintain natural head-to-body proportions. Do not stretch, compress, or resize reference facial features to match an original actor's silhouette. For details not visible in a photograph, use a restrained, plausible continuation consistent with the visible reference. Do not invent distinctive accessories. Do not copy a photograph's background, pose, lighting, or frozen expression. [Character Assignments] PERSON_A, from @image1: The opening performer on the parking lot, corresponding to the source person originally wearing white. PERSON_B, from @image2: The camera operator, visible during BOTH selfie sections. PERSON_C, from @image3: The person who exits the grey car and walks forward. These assignments never change. No identity blending or exchanged outfits. [Environment and Visual Style] Recreate the recognizable setting of @video1: an outdoor roadside parking area near a service station, asphalt ground, trucks and parked cars, a grey car beside a dark SUV, service-station structures, and evening sky. Match the general spatial arrangement so the camera turns and character actions remain coherent. Render a fresh, plausible environment rather than copying source frames. Use plain, unlettered surfaces wherever the source contains signs, captions, logos, or readable text. Casual handheld smartphone footage. Natural evening exposure, realistic skin and fabric, ordinary phone-camera sharpness, mild noise, motion blur during rapid turns, and believable automatic exposure adjustment. No polished commercial look or new cinematic camera moves. [Performance Transfer] Use @video1 to understand the actions and emotions, then have the reference people perform those actions themselves. Transfer the meaning and timing of gestures, head turns, smiles, expressive mouth movements, and reactions. Preserve each reference person's own facial anatomy during every expression. Do not trace the source facial landmarks onto a differently shaped reference face. Do not use the original actors' heads or bodies as fixed templates. Adapt steps, arm reach, balance, and car clearance naturally to the replacement people's physiques. Maintain believable contact with the ground and objects. [Timeline — Approximate Source-Inspired Timing] [00:00–00:06] PERSON_A — OPENING PERFORMANCE Film PERSON_A, built entirely from @image1, on the parking lot. Reproduce the source opening performance: expressive singing-like or shouting-like mouth movements, animated arm gestures, rhythmic body movements, steps, head turns, and emotional intensity. PERSON_A wears the clothing and accessories from @image1, not the source performer's white outfit unless the photograph independently shows the same outfit. Follow the source camera distance, angle, framing, and handheld movement. Keep PERSON_A's characteristic head shape and facial proportions recognizable throughout the performance. [00:06–00:08] PERSON_B — FIRST SELFIE Follow the source camera turn into a close selfie view of the operator. The operator is PERSON_B from @image2, with their own reference face, skull shape, hairstyle, physique, outfit, and accessories. Reproduce the source smile, expressive mouth movements, head motion, and emotional energy without inheriting the original operator's facial anatomy. Use believable close-range phone-lens perspective. Do not add the original bucket hat or black T-shirt unless they are present in @image2. [00:08–00:17] CAMERA TURN AND PERSON_C EXITING THE CAR Reproduce the source's rapid turn toward the parked vehicles and its change in framing or zoom. Show the grey car and nearby dark SUV in a coherent parking layout. PERSON_C, built entirely from @image3, exits the grey car, turns, smiles, and walks forward in the same general sequence and rhythm as the source. Use PERSON_C's reference clothing, headwear, accessories, physique, hairstyle, and facial structure. Adapt the exit naturally to PERSON_C's body dimensions. Show believable footing and hand contact; the person and clothing must clear the door and roof without clipping. Keep the reference identity stable as the face changes angle and moves closer. [00:17–00:20] PERSON_B — FINAL SELFIE Follow the source camera movement back to the operator. Show exactly the same PERSON_B from the first selfie: identical head shape, facial identity, hair, outfit, and accessories. Reproduce the source's final expressive reaction, smile, mouth movements, and handheld framing. End with the same general pacing as the reference. [Population] Only the three designated people may appear. Do not recreate source background pedestrians, bystanders, passengers, or crowds. Keep background walkways and parking spaces unoccupied by additional people. No extra people in windows, reflections, or vehicle interiors. [Strict Text-Free Rendering] Generate clean imagery with NO text at any point. No captions, subtitles, title cards, speech bubbles, emojis, stickers, watermarks, interface elements, readable signs, license-plate characters, or clothing lettering. The source video's text and graphic overlays are excluded from the reference interpretation. They must never appear in the generated scene. Do not create a panel, blurred strip, blank caption box, or other substitute where source text appeared. Render the actual scene across the entire frame. Preserve reference garment colors, materials, and non-text patterns, but omit lettering and typographic logos. [Continuity and Quality] Keep all three identities separate and consistent. Preserve reference head shapes through profiles, smiles, open-mouth expressions, close-ups, and fast camera turns. Natural skin texture and expressions; no pasted-face appearance. No original actors' facial features, source-body silhouettes, identity drift, doubled features, floating hair, distorted hands, or changing accessories. Follow the source camera choreography and approximate pacing without forcing pixel-level alignment with the original frames. [Audio] Generate only natural location ambience and nonverbal human reactions consistent with the scene. No new intelligible dialogue, generated song, or background music. Leave the clip suitable for synchronization with the original soundtrack during post-production. [Final Priority] 1: Faithful, complete character appearances from @image1, @image2, and @image3. 2: Natural anatomy, coherent bodies, and stable identities. 3: Source-inspired actions, emotions, camera choreography, location, and pacing. 4: Clean text-free imagery with no additional people. A newly generated scene performed by the reference people — NOT the original actors with replacement faces.

ПРОМПТ ДЛЯ ВИДЕО (РЕЖИМ - EDIT):

[Video Editing Task] Edit @video1 using @image1, @image2, and @image3 as COMPLETE CHARACTER APPEARANCE references. Fully replace the three designated people, including their identities, hair, physiques, clothing, headwear, footwear, and wearable accessories. Do not perform face-only replacements. Preserve the exact source duration, approximately 20 seconds, vertical 9:16 aspect ratio, original shot sequence, action timing, camera movements, zooms, transitions, and original audio. [Reference Mapping] PERSON_A = the complete person and outfit from @image1. Replace the opening performer originally wearing a white polo shirt, light jeans, and white sneakers. PERSON_B = the complete person and outfit from @image2. Replace the camera operator in BOTH selfie appearances: the middle selfie section and the final selfie section. PERSON_C = the complete person and outfit from @image3. Replace the person who exits the grey car and walks forward after the camera swings toward the parked vehicles. Keep these assignments consistent throughout the entire video. Never exchange identities, outfits, hairstyles, or accessories between the three references. [Full Appearance Replacement — Highest Priority] For each designated character, use their assigned photograph as the source for: - Recognizable facial identity and facial structure. - Apparent age and gender presentation. - Skin tone. - Visible physique and body proportions. - Hair color, hairline, hairstyle, and hair length. - Facial hair, if present. - All visible clothing, including garment cut, fit, layers, colors, patterns, and materials. - Headwear, glasses, jewelry, watches, and other visible wearable accessories. - Footwear, if visible in the reference. Do NOT retain the original video's clothing, hairstyles, hats, or accessories unless the same items are also present in the assigned reference photograph. If the reference person has no headwear, remove the source character's headwear. If the reference includes headwear, reproduce it. Apply the same rule to glasses, jewelry, and other wearable accessories. Where a clothing or body area is not visible in the reference photograph, use a restrained, plausible continuation consistent with the visible appearance. Do not claim to reproduce unseen details. Do not copy the reference photograph's background, pose, lighting, or static expression. Do not introduce handheld objects from the photographs that would interfere with the original actions. Reference appearance determines WHO the person is and WHAT they wear. The source video determines WHAT they do, HOW they react, and HOW the camera moves. [Movement Transfer] Transfer the original movements to the replacement bodies while preserving the exact action timing, screen positions, movement paths, and interactions. Adapt the motion naturally to each reference person's physique. Preserve foot contact with the ground, hand contact with objects, balance, and plausible joint movement. Clothing and accessories must move naturally with the replacement body. Do not preserve the outline of an original garment beneath a different reference outfit. [Approximately 00:00–00:06 — PERSON_A] Fully replace the opening performer with the person from @image1, including their complete reference outfit and accessories. Remove the original white polo, jeans, and sneakers unless those items are actually present in @image1. Preserve every original step, expressive movement, arm gesture, body lean, head turn, gaze direction, and facial reaction. Transfer the source singing or shouting mouth movements and emotional performance to the replacement face. Do not hold the expression from the reference photograph. Preserve the parking lot, trucks, vehicles, evening sky, camera distance, and framing. [Approximately 00:06–00:08 — PERSON_B, First Selfie] Fully replace the operator with the person from @image2. Replace the original black T-shirt and bucket hat with the actual clothing and headwear from @image2. If @image2 contains no hat, show the reference hairstyle without a hat. Preserve the exact camera rotation toward the operator, handheld perspective, lens distortion, face-to-camera distance, smile, mouth movements, and head motion. Keep the operator's appearance consistent during all partially visible and motion-blurred frames. [Approximately 00:08–00:17 — PERSON_C, Car Exit] Preserve the original rapid camera swing and zoom toward the parked vehicles. Fully replace the person exiting the grey car with the person from @image3, including their reference age, gender presentation, physique, hairstyle, outfit, headwear, and wearable accessories. Do not retain the source person's jacket, lower garments, or hairstyle unless they match @image3. Preserve the exact car-exit choreography, stepping motion, turns, smile, walking direction, and timing. Adapt the replacement body and garments naturally to the doorway and seat. Preserve all source contact points and occlusions involving the car door, roof, window frame, hands, and body. No passing through solid surfaces or clothing clipping through the car. Keep the cars, parking layout, and camera movement unchanged. [Approximately 00:17–End — PERSON_B, Final Selfie] Replace the operator again with the SAME complete appearance from @image2. The face, hairstyle, headwear, clothing, and accessories must match the first selfie section exactly. No return of the original bucket hat, T-shirt, or source identity. Preserve the final expressions, mouth movements, camera shake, framing, and ending. [Timing] The time ranges above are approximate navigation guides only. Follow the actual source transitions and action boundaries exactly. Do not retime the video to fit rounded timestamps. [Remove Other People] Keep only the three designated characters, each appearing where their corresponding source character originally appears. Remove all incidental background people throughout the clip, including the passersby in the final selfie section. Do not remove a designated character if they remain visible in the background of another character's shot. Reconstruct the revealed background naturally. Remove visible reflections and shadows belonging to deleted background people. No ghost figures, blurred silhouettes, or flickering repair patches. [Remove All Text] Remove the existing caption overlay, its backing panel, emojis, and any other added graphics throughout the clip. Reconstruct the scene behind the overlay, including moving clothing or body parts when necessary. Do not hide the overlay with cropping, blur, or another panel. No readable text in the final video. Remove visible lettering from garments, accessories, signs, license plates, and watermarks while preserving their underlying colors, materials, shapes, and non-text patterns. This text-removal requirement is the only exception to exact reference outfit reproduction. Do not generate replacement letters or pseudo-text. [Preserve the Source Video] Keep the locations, vehicles, props, evening atmosphere, and lighting unchanged, except for the required character replacements, background-person removal, and text removal. Preserve all original handheld shake, whip pans, zooms, exposure changes, motion blur, transitions, and playback speed. No added stabilization, camera movements, cuts, slow motion, cropping, or reframing. [Facial and Visual Continuity] Preserve each source character's original expressions, blinking, gaze directions, smiles, and lip movements on the assigned replacement identity. Match the source lighting, shadows, exposure, grain, sharpness, and motion blur. Update body shadows and reflections where needed to match the replacement person and outfit. Maintain stable faces, bodies, garments, and accessories across all frames. Respect all foreground occlusions. No identity mixing, original faces reappearing, wardrobe reversion, duplicated accessories, distorted anatomy, oversized heads, mask edges, or floating clothing. [Audio] Preserve the original soundtrack unchanged and synchronized, including vocals, music, and ambience. Do not change voices to match the replacement appearances. No new speech or additional sounds. [Final Result] The same source video with: 1. The opening performer fully replaced by the person and outfit from @image1. 2. The operator fully replaced by the person and outfit from @image2 in BOTH selfie appearances. 3. The person exiting the grey car fully replaced by the person and outfit from @image3. 4. All incidental people removed. 5. All visible text removed. Complete reference appearances; original source performances, camera work, timing, and sound.

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Создаем сцену из человека паука с вашими героями.

2026-09-21T21:29:43.713Z

Нужен аккаунт в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
*Подписка PRO и выше.
Нейросети: Gpt Image 2 для фото, Seedance 2.5 для видео.
Нужно: 2 фотки героев + видео-исходник (1 часть и 2 часть)

1.Создаем карты персонажей.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

Промпт для карты персонажа:

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

RIGHT PANEL (close-up):
A close-up portrait of the person's face, cropped at the shoulders (bust shot). Face front-facing, looking directly at the camera, neutral expression, natural even lighting. Preserve all facial features and details as accurately as possible: eye shape, nose, lips, eyebrows, face shape, eye color, skin tone and texture, hair shape and color, as well as any unique features (moles, freckles, ear shape). High detail, realism, portrait quality.

LEFT PANEL (back view, full body):
The person standing at full height, front view, in their everyday clothing, arms along the body, feet shoulder-width apart. IMPORTANT: the head and face are completely absent — the figure is cropped at the neck/shoulder line, no head. Only the figure from the neck down is shown. Preserve body build, proportions, and clothing.

MIDDLE PANEL (front view, full body):
The same person, back view, in the same clothing, at full height, including the head and hair (viewed from behind). Preserve the hairstyle, body build, and clothing details.

GENERAL REQUIREMENTS:
Unified style, consistent lighting and figure scale across all panels, consistency of appearance and clothing between panels. Clean background, clear vertical dividers between panels. Realistic photographic style, high resolution.

*Со вторым тоже самое.

2.Создаем видео.

Для начала скачиваем 1-ую часть ролика: ССЫЛКА

Дальше переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим EDIT - Качество 720p (480р, чтобы сэкономить, а если мажоры, то 1080р) - Соотношение сторон "авто" - Длительность "авто". Загружаем карты персонажей + исходник видео. После вставляем промпт и вручную удаляем все "@image1", "@image2", "@video1" и заново их прописываем, когда будете вводить "@" там можно будет выбрать нужный файл, так они привяжутся к промпту (если делаете в Syntx.ai, то ничего делать не нужно, они автоматически привяжутся):

ПРОМПТ ДЛЯ 1 ЧАСТИ:

[Video Editing Task] Edit @video1. Replace only the two main characters using @image1 and @image2, including their clothing. Preserve the original video rather than generating a new interpretation. Keep the exact source duration, square aspect ratio, shot order, cuts, camera angles, framing, action timing, original slow-motion segments, and audio. [Character Mapping] @image1 → PERSON_A: The person originally wearing the colorful comic-print zip-up hoodie, initially seated at the cafeteria table, who catches the falling person and the tray. @image2 → PERSON_B: The person originally wearing a white T-shirt and blue jeans, with tousled light hair, who walks through the cafeteria carrying the food tray and slips. These assignments stay fixed throughout the entire clip. Never swap or blend the two identities. [Replace PERSON_A] Replace PERSON_A's appearance with the person in @image1. Preserve the reference facial structure, skin tone, hairline, hairstyle, facial hair if present, and distinctive visible features. Replace the original comic-print hoodie and other visible garments with the clothing shown in @image1. Match the reference garment colors, patterns, cut, and materials. For clothing areas not visible in the photograph, use a simple consistent continuation of the visible outfit. Preserve PERSON_A's exact source performance: seated posture, noticing the approaching person, head turns, supporting the falling person, catching the tray and food, and the final interaction and speech movements. [Replace PERSON_B] Replace PERSON_B's appearance with the person in @image2. Preserve the reference facial structure, skin tone, hairline, hairstyle, facial hair if present, and distinctive visible features. Replace the original white T-shirt and jeans with the clothing shown in @image2. Match the reference garment colors, patterns, cut, and materials. For clothing areas not visible in the photograph, use a simple consistent continuation of the visible outfit. Preserve PERSON_B's exact source performance: walking with the tray, stepping onto the spill, slipping, losing balance, dropping the tray, being caught, and looking at PERSON_A at the end. [Preserve the Original Choreography] Do not change either character's screen position, movement path, pose timing, gaze direction, or physical interaction. Keep the original body blocking and contact points so the supporting arm, falling body, hands, and tray remain correctly connected. Adapt the replacement clothing naturally to the existing movements, with believable folds and fabric deformation. Preserve the exact timing and trajectories of the airborne food and tray, including every original slow-motion moment. Keep all original reaction shots, close-ups, floor inserts, and the final shot of the two characters together with the caught tray. Do not invent new gestures, expressions, dialogue, falls, or catching actions. Transfer each source facial expression and mouth movement to the corresponding replacement face; do not freeze the expression from the reference photograph. [Consistency and Occlusion] Apply both replacements consistently in every shot, including profiles, rear views, partially hidden faces, and motion-blurred frames. Preserve correct foreground occlusion: hands, arms, the other person, the tray, and flying food must obscure the replacement face or clothing exactly when they do in the source. Match the original lighting, shadows, perspective, depth of field, grain, and motion blur. Blend faces naturally around the hairline, ears, jaw, and neck. Maintain stable identities and outfits across cuts. No face flicker, identity drift, original faces reappearing, duplicated limbs, fused hands, floating clothing, or visible mask edges. [Unchanged Elements] Leave all background people unchanged. Preserve the cafeteria, tables, chairs, floor, spill, food, tray, and every other prop. Preserve the original audio unchanged, including dialogue, ambience, music if present, and synchronization. Do not add or remove shots, change playback speed, crop, reframe, or alter camera movement. No new text, captions, logos, or watermarks. Final result: the same cafeteria clip, with PERSON_A replaced by the appearance and clothing from @image1, and PERSON_B replaced by the appearance and clothing from @image2. Everything else remains as in @video1.

Теперь скачиваем 2-ую часть ролика: ССЫЛКА

Также в Раздел Видео - Выбираем Seedance 2.5, настройки такие же, только теперь загружаем 2-ую часть исходника + фото первого человека:

ПРОМПТ ДЛЯ 2 ЧАСТИ:

[Video Editing Task] Edit @video1 using @image1 as the identity reference for the recurring main character. Replace only the main character's facial identity and hair throughout the entire clip. Preserve the source body, clothing, actions, facial performances, locations, props, editing, and audio. Keep the exact source duration, approximately 8 seconds, original 1:1 square aspect ratio, and every original cut. Do not rebuild the montage from scratch. [Identity Reference] @image1 = the replacement identity. Match the person's recognizable facial structure, skin tone, eye color, hairline, hairstyle, and facial hair if present. Do not copy the photograph's clothing, background, pose, lighting, or static expression. Fit the replacement identity naturally to the source head position, perspective, and body proportions. Preserve the source head's approximate scale within the frame; no oversized head or altered body silhouette. [TARGET — Recurring Main Character Only] Replace the principal character wearing the dark blazer, light shirt, and diagonally striped tie in ALL eight shots. Do not replace other people, including the foreground person in the dark-room shot or the background person near the doorway. [Shot Coverage — Follow the Existing Source Cuts] 1. OUTDOORS NEAR THE FENCE AND TRACTOR: Replace the main character's face and hair during the quick backward head turn. Preserve the original alert expression, gaze, outdoor lighting, and motion blur. 2. DARK INTERIOR WITH ANOTHER PERSON IN THE FOREGROUND: Replace the main character's face and hair while preserving the smile and head movement. Keep the foreground person's head unchanged and preserve its original overlap with the main character. 3. BRIGHT ROOM WITH A BLACK MUG: Replace the main character's face and hair. Preserve the original smirk, gaze, mug, hand placement, and window lighting. 4. CORRIDOR, LOOKING BACK OVER THE SHOULDER: Replace the face and hair throughout the turn. Preserve the original shoulder position, walking movement, pleased smile, and camera framing. 5. DOORWAY: Replace the main character's face and hair during the half-turn and smile. Keep the background person, doorway, and all movement unchanged. 6. DARK PROFILE WITH A GLASS CONTAINER: Replace the main character's facial identity and hair in profile, respecting the source silhouette and very low exposure. Keep the replacement naturally dark. Do not brighten the face or invent frontal facial detail in a silhouette shot. Preserve the glass container and hands. 7. GLASS DOOR OR BLINDS: Replace the main character's face and hair during the smiling shoulder-dance movement. Preserve the exact rhythm, body motion, outfit, and background. 8. SEATED AT THE DESK: Replace the main character's face and hair. Preserve the original pointing gesture toward the camera, mischievous squint, smile, desk, props, and final-frame timing. [Preserve Clothing and Body] Keep ALL source garments exactly as shown in each shot: dark blazer, light shirt, diagonally striped tie, trousers where visible, and any existing accessories. Do not transfer any clothing from @image1. Do not change the source torso, limbs, hands, body proportions, posture, or movement. Only blend the replacement head naturally into the existing neck and collar boundary. [Preserve the Source Performance] Transfer each shot's original expression onto the replacement identity: smiles, smirks, squints, alert looks, and gaze changes. Preserve original head rotations, eye directions, blinking, mouth movements, and action timing. Do not impose a single expression across all shots. Do not make the character look toward the camera unless they do so in the source. [Temporal Consistency and Compositing] Maintain the same replacement identity across every cut, angle, and lighting condition. No original face reappearing during turns or partially obscured frames. Match the source lighting, shadows, exposure, color grade, sharpness, grain, depth of field, and motion blur. Preserve realistic skin texture. Respect all original occlusions by foreground people, objects, doorframes, and hands. Keep clean, natural boundaries around the hair, ears, jaw, and neck. No face flicker, identity mixing, doubled features, floating hair, mask edges, stretched faces, or sudden head-size changes. [Unchanged Elements] Preserve the exact shot order, cut points, playback speed, camera movements, composition, locations, and props. Preserve the original audio track unchanged and synchronized. No new scenes, additional gestures, dialogue, music, captions, text, logos, or watermarks. Final result: the same eight-shot source montage with the recurring main character's identity replaced by @image1, while the original clothing, body, performances, editing, and sound remain unchanged.

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Создаем видео: "Когда купил новую машину и показываешь её друзьям..."

2026-09-20T15:54:19.568Z

Нужен аккаунт в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
*Подписка PRO и выше.
Нейросети: Gpt Image 2 для фото, Seedance 2.5 для видео.
Нужно: ваше фото, фото машины (желательно с разных сторон).

1.Создаем карту персонажа и машины.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

ПРОМПТ ДЛЯ КАРТЫ ПЕРСОНАЖА:

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

ПРОМПТ ДЛЯ КАРТЫ МАШИНЫ:

Take the car's appearance 100% from the uploaded reference images (front view and rear view of the same car) and preserve absolutely every detail exactly as it is: body shape, color, paint finish, grille, headlights, taillights, bumpers, wheels, rims, badges, trim and proportions — whatever car it is, nothing added, nothing removed, nothing redesigned. Identify the car's brand and model from the badges and lettering visible in the reference images.
Create a premium automotive spec card, 16:9 horizontal layout, clean professional presentation design:
— TWO LARGE MAIN PANELS side by side. LEFT PANEL: the car in a FRONT three-quarter view, camera positioned slightly above eye level so the roof of the car is clearly visible. RIGHT PANEL: the car from the same slightly elevated camera height, but in a REAR three-quarter view from the opposite side, roof visible as well. In both panels the car is fully in frame, perfectly lit studio-style, identical scale in both panels.
— ONE SMALL SQUARE PANEL centered at the bottom between the two main panels: a macro close-up of the car's brand badge/logo exactly as it appears on the car in the reference images, sharp and detailed, filling the square.
— At the TOP CENTER of the card: the car's name written as clean elegant text — use the exact brand and model taken from the reference images (from its badges and lettering), in a modern premium sans-serif font, large and clearly legible, correctly spelled.
Style: high-end car configurator / brochure aesthetic, neutral dark studio background with soft gradient lighting behind the car in both panels, subtle floor reflections under the wheels, thin minimal panel borders or soft separation between panels, cohesive color scheme matching the car's paint color as an accent. Photorealistic renders of the car with real physical texture: accurate reflections on paint and glass, correct shadows under the body.
Do not change the car's design, color or proportions — it must look exactly like the reference images in every panel. No people, no other cars, no extra text beyond the car's name at the top, no watermarks, no fictional badges, no misspelled lettering, no distorted logo in the badge close-up.

2.Создаем видео.

Есть два варианта:
-Отправить промпт с музыкой
-Отправить промпт без музыки
*Разницы почти нет, но я отправлял с музыкой.
Мелодия для промпта: ССЫЛКА
Музыка для монтажа: ССЫЛКА

Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - Качество 720p (480р, чтобы сэкономить, а если мажоры, то 1080р) - Соотношение сторон 16:9

Длительность 20 сек (если отправляете вместе с мелодией, то выбираете "AUTO").

Загружаем карту персонажа + КАРТУ МАШИНЫ (также мелодию, при желании). После вставляем промпт и вручную удаляем все "@image1", "@image2" и заново их прописываем, когда будете вводить "@" там можно будет выбрать нужный файл, так они привяжутся к промпту (если делаете в Syntx.ai, то ничего делать не нужно, они автоматически привяжутся):

ПРОМПТ ДЛЯ ВИДЕО.

[REFERENCES] @image1 — character reference sheet. Preserve the exact person's face, apparent age, gender presentation, hair, physique, clothing, footwear, and accessories throughout. Use only the person and outfit from this sheet, NOT its background, lighting, poses, or panel layout. @image2 — vehicle reference sheet. Preserve the exact car, paint color, proportions, wheels, lights, glass, mirrors, handles, grille, badges, and body details. Use only the vehicle design, NOT the reference background or studio lighting. @audio1 — original soundtrack and rhythmic editing reference. Use this exact audio from the beginning, at its original speed and pitch, without replacement music. The reference sheets show different views of ONE person and ONE car. Do not reproduce their panels or create duplicates. [CONCEPT AND LOCATION] A cinematic car-owner showcase on an EMPTY OUTDOOR PARKING LOT IN RUSSIA ON AN AUTUMN NIGHT. This outdoor location is mandatory in every shot. Dark open sky, weathered asphalt, faded parking lines, scattered yellow and brown leaves, bare trees, ordinary tall streetlamps, and distant subdued apartment-building silhouettes. Slightly damp asphalt reflects the streetlights. No other parked cars or pedestrians. Streetlamps are the only environmental light sources. The car's actual lights may also be visible. Preserve nighttime darkness between illuminated areas. Paint and glass catch narrow reflections from real streetlamps and surrounding trees. NO studio, showroom, cyclorama, seamless grey background, indoor ceiling, daylight, or featureless floor. Even close-ups must retain outdoor context through asphalt, leaves, streetlamp bokeh, or physically consistent reflections. [DURATION AND EDIT] Match @audio1, approximately 19.51 seconds. One opening shot until approximately 00:04.1, then EXACTLY TWELVE presentation shots. Use the timeline below as approximate guidance and align cuts with prominent accents in @audio1. No extra shots or end card. The opening looks like authentic handheld smartphone footage. At the first detail shot, switch decisively to polished professional automotive cinematography while keeping the same location, night, person, outfit, and car. If the output requires 20 seconds, hold the final image after the soundtrack finishes; do not stretch or loop @audio1. [00:00–00:04.1 | HANDHELD PHONE OPENING] Realistic nighttime smartphone footage recorded by someone standing nearby. Slight irregular hand tremor, modest low-light noise, natural motion blur, and a small reactive framing adjustment. No smooth commercial orbit or artificial slow motion in this opening. The person from @image1 stands in the foreground with arms crossed. Their torso is turned approximately 30–45 degrees away from the camera, while their head turns toward the lens. Serious model-like expression, closed mouth, relaxed confidence. Maintain the exact reference clothing. The car from @image2 is already moving quickly in the background at the start. An adult woman is seated at the correct driver's position, visible only naturally through the windows when the angle allows. She remains inside and is not a second presenter. Show a fast approach on a diagonal behind the foreground person, followed by one decisive sideways drift resembling a cinematic handbrake turn. The rear swings outward while the whole vehicle continues translating across the asphalt. The car must travel along a coherent path, NOT rotate in place. Maintain wheel contact with the road, plausible steering, weight transfer, and continuous momentum. Restrained tire haze trails from the road-contact area; do not hide the vehicle behind smoke. The drift finishes behind the person, with the car settling into a side-dominant three-quarter view. The stopping position appears almost touching the person, with only a hairline gap visible in the final screen composition. Create this near-contact illusion through compositing and compressed perspective, not an actual close pass. No collision, overlap, or contact. The person does not step away or uncross their arms. A small handheld camera reaction emphasizes the sudden arrival. Finish the stop before the musical cut around 00:04.1. No supernatural sliding, instant direction changes, wheel skating, or rotating-car display-platform effect. [00:04.1–00:05.4 | 01: FRONT WHEEL] Switch to professional cinema-camera footage. Low stabilized lateral track past the stationary front wheel, with strong parallax between the tire, bodywork, and distant streetlights. Damp asphalt and leaves remain visible. The person is already crouched beside the wheel, recognizable face at the frame edge. One index finger points precisely toward the visible rim center without covering it. Their gaze moves from the rim to the lens. Mouth closed, eyebrows subtly lifted. [00:05.4–00:06.7 | 02: FRONT EMBLEM] Close diagonal camera slide across the front grille toward the exact emblem. Streetlamp reflections move naturally across the badge as the viewing angle changes. The person's face is already close beside the emblem. An open palm sits just below it like a presentation stand. Eyes glance toward the badge, then toward the camera with quiet satisfaction. No lip movement. [00:06.7–00:08.0 | 03: HEADLIGHT] Short curved tracking move around the front corner, tracing the real headlight shape. Visible perspective change across the lamp, fender, and background lights; not a digital zoom. The person leans into the far edge of the composition. Their fingertip indicates the actual headlight contour, stopping beside the lens rather than covering it. Eyes follow that contour. Closed neutral mouth. [00:08.0–00:09.3 | 04: SIDE MIRROR] Close stabilized arc from the mirror's front edge toward its outer side. The mirror remains sharp while bare-tree silhouettes and streetlights shift behind it. The person's face is visible beside the mirror. An open hand presents the housing from below. They glance from the mirror to the lens with a small eyebrow lift. Do not wave or speak. [00:09.3–00:10.6 | 05: DOOR HANDLE] Smooth lateral dolly along the door, ending with the actual handle prominent. Moving streetlamp reflections reveal the paint and panel contour. The person stands beside the door, face visible near the upper frame edge. Their index finger points directly beside the handle, not toward an unrelated panel. Keep the door closed and the feature unobstructed. [00:10.6–00:11.9 | 06: SIDE SILHOUETTE] Low professional tracking move parallel to the car, revealing its side proportions with clear foreground-to-background parallax. Include parking markings and autumn leaves. The person stands near one end of the car, occupying a small portion of the frame. Both open palms present the vehicle's length. Their gaze follows the body line, then returns to the lens. Calm face, pleased eyes, closed mouth. [00:11.9–00:13.2 | 07: REAR WHEEL] Short low arc around the stationary rear wheel and fender. A nearby blurred leaf provides foreground depth while the wheel remains clear. The person is already crouched at the opposite frame edge. Point precisely toward a visible wheel spoke with one hand. Keep the face recognizable and the gesture anatomically natural. [00:13.2–00:14.5 | 08: TAILLIGHT] Close tracking pass around the rear corner, following the actual taillight and revealing part of the adjoining bodywork. Rich but realistic red reflections on damp surfaces. The person appears beside the light with an open palm immediately below it. Their eyes turn toward the taillight, then toward the viewer. No grin, no teeth, no moving lips. [00:14.5–00:15.8 | 09: REAR EMBLEM] Short diagonal dolly toward the exact rear emblem. Keep both the emblem and the person's nearby face readable. The person tilts their head slightly toward the badge and presents it primarily with eye direction and a subtle eyebrow lift. Hands may remain out of frame. Preserve reference lettering; do not invent or substitute branding. [00:15.8–00:17.1 | 10: LOWER REAR DETAIL] Very low lateral camera slide along an actual lower rear design feature visible in @image2. Keep the outdoor asphalt and scattered leaves visible. The person is already crouched at the edge, with part of their face included. Point directly beside the selected visible component. Show exhaust outlets or a diffuser only if the referenced car has them. [00:17.1–00:18.3 | 11: HOOD AND ROOFLINE] A short elevated diagonal crane move over the front quarter, revealing the hood, windshield, and roofline. Maintain smooth professional motion and moving streetlamp reflections. No impossible flight through glass or bodywork. The person stands beside the car, looks up toward the camera, and presents the visible hood line with one open palm. Keep the same face, clothes, and restrained expression. [00:18.3–00:19.51 | 12: FINAL HERO SHOT] Full car in a dramatic three-quarter view on the same empty autumn parking lot. Smooth low pullback combined with a restrained arc, showing the vehicle against streetlights and bare trees. The person stands beside the car without blocking it. Present the vehicle with both open palms, then hold the pose. Eyes show unmistakable satisfaction, but the mouth remains closed and neutral. Finish on this composition with the end of @audio1. [EXPRESSION AND SILENCE — MANDATORY] The presenter NEVER speaks, lip-syncs, mouths words, laughs, or shows teeth. Keep lips gently closed and relaxed in every shot. No visible smile or grin. Express happiness and pride through attentive eyes, slight eyebrow lifts, confident eye contact, and subtle head movements. Use believable gaze shifts between the actual featured detail and the lens, not exaggerated pupil movements or crossed eyes. All presentation is nonverbal. Do not interpret any gesture as an invitation to generate dialogue. [GESTURES AND FRAMING] The same recognizable presenter appears in every shot. A detached hand alone is insufficient. Place their face beside the feature, near a frame edge, or farther back within a readable depth of field. Each shot begins with the person already correctly positioned. One clear gesture per shot. Every pointing finger must indicate the specific featured component visibly inside that same frame. Keep fingertips beside the feature rather than covering it. No pointing offscreen, at empty space, or at unrelated parts. [CAMERA AND CONTINUITY] Opening: handheld smartphone realism. All twelve subsequent shots: stabilized professional camera movement with visible spatial parallax, controlled focus, natural motion blur, and cinematic reflections. No static slideshow, repeated digital zooms, frozen images, or identical camera moves. The car remains stationary after the opening. All wheels stop rotating. Maintain the same outdoor parking lot, nighttime illumination, exact vehicle, reference outfit, and presenter identity. The driver remains inside if visible; she never replaces or duplicates the presenter. Use direct beat-aligned cuts. No dissolves, morphing, costume changes, or relocation to a studio. [SOUND AND EXCLUSIONS] @audio1 is the sole soundtrack. Preserve its existing vocals and music, but do not animate the presenter singing or speaking along with them. No added dialogue, voice-over, engine sounds, tire sounds, or replacement music. No subtitles, text overlays, watermarks, additional cars, bystanders, invented vehicle components, altered emblems, duplicated people, distorted hands, body intersections, or collisions.

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Ролик, Когда спрашивает мама, "Как у тебя дела?"

2026-09-19T14:51:58.928Z

Нужен аккаунт в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
*Подписка PRO и выше.
Нейросети: Gpt Image 2 для фото, Seedance 2.5 для видео.
Нужно: только ваше фото.

1.Создаем карту персонажа.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

Промпт для карты персонажа

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

2.Создаем видео.
Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - Качество 720p (480р, чтобы сэкономить, а если мажоры, то 1080р) - Соотношение сторон любое ( у меня 4:3) - Длительность 15 сек. Загружаем карту персонажа. После вставляем промпт и вручную удаляем все "@image1", и заново их прописываем, когда будете вводить "@" там можно будет выбрать нужный файл, так они привяжутся к промпту (если делаете в Syntx.ai, то ничего делать не нужно, они автоматически привяжутся):

ПРОМПТ ДЛЯ ВИДЕО:

Create a 15-second photorealistic cinematic montage consisting of exactly ten shots, with direct cuts at the specified times. Only one reference image is used: @image1. No video or depth-map reference is required. [Identity] Use the same recognizable person from @image1 in every shot. Preserve their facial structure, apparent age, gender presentation, skin tone, hair color, hair length, hairstyle, and existing facial hair. This prompt applies to any gender. Fit the specified costumes naturally to the referenced person. Do not copy the photograph's clothing, background, lighting, or expression. [Central Emotional Direction — Applies to Every Shot] The recurring expression is a bittersweet, reassuring smile paired with a thumbs-up. The person tries to appear positive during a difficult day. The smile is gentle and slightly hesitant rather than openly cheerful. Their eyes remain thoughtful and subdued. Communicate this through subtle facial acting: slightly raised inner eyebrows, a small asymmetric smile, a brief pause before smiling, lightly pressed lips, a slow blink, a quiet breath, or a moment when the smile softens. Use different combinations in different shots. Keep the performance natural and restrained. Visible tears are unnecessary. The contrast between the tentative smile and the subdued eyes should carry the emotion. Every thumbs-up has one extended thumb and four naturally folded fingers. [00:00–00:01.2 — A BRAVE LITTLE SMILE] Close-up at eye level in a bright outdoor poolside setting with a softly blurred background. The person wears a pale baseball cap and a plain white T-shirt. Their face is slightly flushed, with faint pink cosmetic smudges and a subtle sheen of moisture. Begin with a subdued expression and slightly raised inner eyebrows. They form a small reassuring smile and raise their right thumb beside the face. The mouth corners lift gently, while the eyes remain wistful. Hold the gesture as if quietly saying, “I’m doing my best.” No laughter or exuberant excitement. Bright natural daylight. Static camera. [00:01.2–00:02.4 — QUIET ENCOURAGEMENT] Direct cut to a waist-up shot in an interior with blue tiled walls. Grey button-front shirt beneath a dark charcoal cardigan. Clean skin and natural grooming. Looking toward someone just beside the camera, the person raises BOTH thumbs at chest height and gives a small encouraging nod. A careful closed-mouth smile forms. Their eyes remain thoughtful, and the lips briefly press together before relaxing. The gesture feels considerate and reassuring rather than celebratory. Soft indoor light. Static eye-level camera. [00:02.4–00:03.6 — KEEPING GOING] Direct cut to a tight, slightly elevated view inside a rugged vehicle cabin. Dusty grey jacket. A light sheen of perspiration and a few dusty marks on the face. Warm sunlight enters the darker cabin from one side. With the torso angled away, the person glances back and briefly raises one thumb. They pause for a fraction of a second, then form a small, uneven smile. Their eyelids remain slightly lowered, and their expression is composed but tired. The gesture suggests quiet determination. Subtle vehicle vibration. [00:03.6–00:04.8 — A HESITANT SMILE] Direct cut to a close-up in a crowded, warmly lit interior. Background people remain indistinct. Dark athletic zip jacket with pale stripes. Slight perspiration and natural skin texture. The person presses their lips together briefly, then lifts the corners into a tentative smile while raising one thumb. Their inner eyebrows remain slightly raised. They hold the gaze of someone nearby, then let the smile soften before gently restoring it. The expression is sincere, reserved, and reassuring. Static camera, shallow depth of field. [00:04.8–00:06.3 — RESERVED REASSURANCE] Direct cut to a medium three-quarter shot in a dark industrial room with drifting steam and cool blue backlight. Dark utility jacket, practical vest, and dark gloves. Slight moisture on the skin. Preserve the person's reference age and hair color. Beginning partly turned away, the person looks back over their shoulder and raises a gloved thumb. A small asymmetric smile appears. Their eyes remain serious and reflective rather than triumphant. They release a quiet breath while holding the gesture. Slightly low camera angle. [00:06.3–00:07.5 — MAKING THE BEST OF THE MOMENT] Direct cut to a waist-up office shot with a bright window and blinds behind the person. Tailored three-piece business suit, shirt, and tie. One small pigeon perches on the head, another on one shoulder, and a small blue-toned bird on the other. The person raises BOTH thumbs with a polite, slightly hesitant smile. Their inner eyebrows lift gently, and they give a slow blink. The smile remains modest, while the eyes look thoughtful and somewhat tired. The gentle humor comes from their composed response to the birds, not from exaggerated happiness. Birds make small natural balancing movements. Soft daylight. Static camera. [00:07.5–00:08.7 — A GENTLE GESTURE FROM BED] Direct cut to a close-up of the person reclining against white pillows in a clean hospital room. Hospital gown, white supportive neck collar, and a white medical support around the forearm, leaving the hand and thumb free. The person looks toward a visitor beside the camera and slowly forms a tender, tentative smile. Their eyes remain soft and subdued. They lift the supported forearm slightly and give a small thumbs-up. The expression communicates warmth and quiet reassurance, not exuberance. Keep the head comfortably supported with minimal neck movement. Cool clinical lighting. [00:08.7–00:10.0 — A SMILE FOR SOMEONE WATCHING] Direct cut to a medium shot of the person riding as a passenger in an open red convertible on a sunny city street. Brown tweed jacket, white shirt, and narrow red tie. Their reference hair moves naturally in the wind. The camera faces them from near the windshield and moves with the car. The street passes behind them. They lift their torso slightly and raise BOTH thumbs with a deliberately encouraging smile. The mouth smiles more noticeably than the eyes. After a brief blink, the smile softens into a smaller, more natural expression. No cheering or open-mouthed laughter. Bright surroundings contrast with their quiet, reflective mood. [00:10.0–00:11.8 — A WISTFUL OVER-THE-SHOULDER SMILE] Direct cut to a close three-quarter shot in a softly lit domestic interior. Simple pink thin-strapped top. Preserve the reference hairstyle and hair length. With the body partly turned away, the person looks back over the shoulder. They offer a small, tender smile and a gentle thumbs-up near the shoulder. Their eyes are soft and wistful rather than mischievous. Hold the reassuring gesture briefly. As their gaze begins to drop, the smile becomes less pronounced. Warm light, softly blurred background, static camera. [00:11.8–00:15.0 — TRYING AGAIN] Direct cut to a medium side view beside a quiet country road and a vivid yellow flowering field. Brown tweed jacket, white shirt, and red tie. Bright sunlight and a light breeze. One arm extends toward the road with the thumb raised in a hitchhiking gesture. At first, the person gives an approaching offscreen vehicle a small hopeful smile. Their gaze follows it as it passes without stopping. The smile gradually softens. They lower their eyes briefly and release a quiet breath. Then they look along the road again and gently lift the mouth corners, keeping the thumb raised. End on a modest, hopeful smile with thoughtful eyes. No exaggerated grimaces. Static camera. [Rendering and Editing] One consistent identity across all ten shots. Expressions change naturally without altering facial proportions. Maintain photographic skin texture, believable fabric, realistic hands, and physically coherent props. Preserve the person's own hair and existing facial hair. Costumes, locations, and temporary surface details change only at the direct cuts. Each scene is a single continuous shot. No additional shots, dissolves, costume morphing, or slow motion. The emotional continuity comes from the same subtle contrast: a reassuring smile and thumbs-up, accompanied by thoughtful, subdued eyes. Keep expressions restrained. Avoid exaggerated crying, theatrical reactions, broad celebratory grins, or cheerful advertising-style performances. No dialogue, music, captions, logos, or watermarks. Silent output for later soundtrack editing.

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Вы в роли морского гиганта!

2026-09-18T15:23:48.983Z

Нужен аккаунт в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
*Подписка PRO и выше.
Нейросети: Gpt Image 2 или Grok Imagine 2.0 для фото, Seedance 2.5 для видео.
Нужно: 2 фото пассажиров, 1 фото "гиганта"

1.Создаем карты персонажей.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

Промпт для карт персонажей:

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

*с остальными тоже самое.

2.Создаем видео.
Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - Качество 720p (480р, чтобы сэкономить, а если мажоры, то 1080р) - Соотношение сторон любое ( у меня 9:16) - Длительность 17 сек. Загружаем КАРТЫ ПЕРСОНАЖЕ (2 пассажира и гиганта). После вставляем промпт и вручную удаляем все "@image1", "@image2", "@image3", и заново их прописываем, когда будете вводить "@" там можно будет выбрать нужный файл, так они привяжутся к промпту (если делаете в Syntx.ai, то они автоматически привяжутся):

ПРОМПТ ДЛЯ ВИДЕО:

[Global Parameters] Duration: 17 seconds. Aspect ratio: 9:16 vertical. One continuous photorealistic action-camera shot. Inputs: THREE CHARACTER REFERENCE IMAGES ONLY. No video reference. PRIORITY ORDER: 1. Strictly transverse camera direction: directly across the vessel, never diagonally. 2. Camera permanently bolted to the hull: boat structures stay fixed in the frame while the sea horizon tilts strongly. 3. Fast diagonal giant breach: head visibly above water by 00:03, sideways landing completed by 00:05.5. 4. Three distinct reference identities, consistent clothing and scale. 5. Natural frightened reactions without eye contact with the lens. [Creatives Description] PERSON_A = the person in @image1. Preserve their apparent age, gender presentation, facial features, hairstyle, skin tone, physique, and EXACT clothing shown in the reference. Normal human scale. PERSON_B = the person in @image2. Preserve their apparent age, gender presentation, facial features, hairstyle, skin tone, physique, and EXACT clothing shown in the reference. Normal human scale. GIANT_C = the person in @image3. Preserve their apparent age, gender presentation, facial features, hairstyle, skin tone, physique, and EXACT clothing shown in the reference. Enlarge the whole person proportionally to approximately twelve times ordinary human height. Their head alone is roughly the height of a normal standing adult. Their clothing is drenched from the beginning. Keep all three identities separate. No transferred facial features, duplicated faces, exchanged garments, or additional people. Do not add swimming goggles or equipment absent from the images. [One-sentence Summary] A camera bolted rigidly to the side of a violently rocking full-size wooden vessel looks directly across its deck at two people chatting, while a colossal human breaches diagonally from the water beyond the opposite side, crashes onto their flank, sends a wave across the deck, and resurfaces to watch the terrified passengers. [Camera and Scene Geometry — Mandatory Throughout] VESSEL: A full-size wooden yacht-style vessel, approximately 11 metres long and 4 metres wide, with a broad walkable deck, substantial hull, and solid wooden bulwarks. No mast, sails, or overhead rigging. Enough deck space for the two passengers to stand and move comfortably. Not a rowboat, canoe, dinghy, or miniature. CAMERA POSITION: A wide-angle action camera is permanently bolted to a rigid mounting on the inboard face of the STARBOARD side at MIDSHIP, approximately 0.8 metres above deck level. CAMERA DIRECTION: The lens points directly toward the PORT side, ACROSS THE WIDTH of the deck. The optical axis is perpendicular to the vessel's longitudinal centreline: a strict 90-degree transverse view. Do not look toward the bow or stern. Do not look diagonally along a side passage. COMPOSITION: The opposite port-side bulwark runs LEFT TO RIGHT across the frame, approximately parallel to the bottom image edge. It does NOT form a long diagonal receding into the distance. The deck fills the lower portion of the image. The two people occupy the middle foreground. Open water and sky remain clearly visible beyond the opposite side. Two short, sturdy, waist-high wooden handrails extend inward from the opposite bulwark toward the camera, stopping around the middle of the deck. They border the central space where the two people stand. Each passenger uses their own adjacent internal rail, not the outer bulwark. Use wide-angle GoPro-like optics, mild edge distortion, and deep focus. Frame both standing faces and final crouching positions without moving the camera. RIGID ATTACHMENT AND HORIZON: The camera and hull are ONE rigid moving object. The camera never rotates or translates independently of the vessel. The deck, opposite bulwark, and handrails remain locked in position and orientation relative to the image edges throughout the shot. The SEA HORIZON is not stabilized. It repeatedly tilts strongly relative to the boat, swings through level, and tilts the opposite way. Use combined hull roll and pitch in irregular crossing waves, producing pronounced alternating horizon slopes of approximately 15–25 degrees in this transverse camera view, together with visible vertical movement of the sea. These are heavy, smooth, irregular wave-driven oscillations, approximately one full back-and-forth cycle every 2–3 seconds, not rapid handheld jitter. The boat is the stable reference INSIDE THE IMAGE. The surrounding sea and sky move relative to it. [Specific Plot Description — Timestamped Storyboard] [00:00–00:02.5] CONVERSATION ON AN ALREADY ROCKING DECK Begin immediately with the established transverse camera composition and strong hull movement. No calm establishing shot. PERSON_A stands screen-left and PERSON_B screen-right between the internal rails. Both face slightly toward each other in three-quarter profile, chatting casually. Their eyes stay on each other, never on the camera. Their conversation is unintelligible beneath the wind and waves. Each holds their adjacent wooden rail with ONE hand. Their bodies continuously compensate for the rocking with bent knees, shifting weight, and small shoulder movements. The deck and rails remain fixed relative to the lens while the sea horizon visibly swings from a strong slope in one direction toward the opposite slope. [00:02.5–00:05.5] ONE FAST DIAGONAL BREACH Beyond the opposite PORT side, several metres clear of the hull, GIANT_C bursts diagonally from the sea. At approximately 00:02.5, the head breaks the surface. By EXACTLY 00:03, the entire head is clearly above water, already moving UPWARD AND SIDEWAYS, with the shoulders following immediately. The giant travels from screen-right toward screen-left through the open-water background. Their body emerges at an oblique angle, approximately 35–45 degrees above the water surface, NOT vertically upright. Head, shoulders, torso, and upper legs follow the same continuous momentum-driven arc. Both arms remain straight and pressed firmly against the sides, hands beside the thighs. The body is partly side-on, with the chest and face angled enough toward the vessel for the reference identity to remain recognizable. The giant's expression is stony and cold. Their eyes track the passengers briefly DURING the movement, without pausing. Water streams from the exact reference clothing. The giant's torso dwarfs the passengers and vessel, with consistent human proportions. At the first eruption of water, both passengers snap their heads toward the giant, then turn their shoulders toward the opposite side while retaining their grips. Their expressions change immediately to fear. The giant follows a brief heavy arc, rolls onto their flank, and falls directly back into the sea. Complete the entire emergence and sideways landing by 00:05.5. No upright standing pose, no elevator-like rise, no hovering, no pause at the apex, no slow motion. The giant lands on their SIDE clear of the vessel. Their broad sideways impact displaces water toward the port side. [00:05.5–00:09.5] THE WAVE STRIKES AND THE HULL REBOUNDS Continue from that single impact. Do not repeat the fall. A large splash erupts at the landing point. A broad visible ridge of displaced water travels from the giant toward the vessel. Both passengers immediately grip their respective internal rails with BOTH hands and bend their knees. The wave reaches the opposite port side, lifts and shoves the hull, and spills a shallow surge of foamy seawater over the bulwark. The impact produces a much stronger combined rolling and pitching impulse, followed by two or three visible rebounds. The sea horizon swings sharply relative to the fixed boat structures. The camera remains absolutely rigid against the hull. Do not stabilize the horizon during the impact. The passengers stagger against their grips, shoulders and knees absorbing the force. Neither falls overboard. Water sweeps across the deck, drenching their exact reference outfits and hair. Their clothing darkens naturally with wetness. Spray briefly spatters the lens but does not hide the entire action. Water flows toward deck-edge drainage openings instead of appearing and disappearing instantly. The vessel remains intact and afloat. [00:09.5–00:14] THE ENORMOUS HEAD SWIMS ALONGSIDE GIANT_C resurfaces beyond the port side, toward screen-left. Only the enormous head remains above the surface. Neck, shoulders, arms, and torso stay submerged. Preserve exactly the same giant scale and reference face. The head glides slowly from screen-left toward screen-right along a short curved path beside the vessel, remaining within the transverse camera's view. Show clear sideways travel relative to the opposite bulwark, a broad trailing wake, and ripples displaced by the submerged body. This is continuous swimming, not a stationary head rising vertically. GIANT_C watches the two passengers with a silent predatory stare, closed mouth, no smile. The threatening effect comes from immense size, proximity, and unwavering attention. The passengers follow the head with their eyes and turn their heads along its path. Both remain drenched and grip the rails with two hands. Strong hull motion continues. The background sea horizon keeps tilting relative to the boat; do not let the camera settle into a level stabilized view. [00:14–00:17] CROUCHING AND AVERTING THEIR FACES Both passengers recoil from the giant and lower into frightened crouches without releasing their two-handed grips. They turn their faces away from the threat toward the interior of the deck. This reveals their wet faces in three-quarter view, but their eyes look DOWN toward the deck, toward their hands, or sideways toward each other. They never meet the camera lens with their eyes. Do not create a posed front-facing reaction shot. Show wide frightened eyes, tense brows, trembling lips, hunched shoulders, and strained breathing. End with both drenched passengers crouched at the internal rails, the giant's enormous watchful head visible beyond the opposite side, and the sea horizon still swinging around the boat-relative frame. No attack before the end. [Global Supplement] CAMERA CHECK: Strictly across the width of the vessel. Opposite bulwark spans left to right. No diagonal passage leading toward the bow or stern. Camera mounting never changes. Boat geometry stays fixed relative to the image. The world horizon visibly tilts relative to the boat. If the people and background were hidden, the visible deck and handrails would remain almost motionless inside the frame. If the boat were hidden, the sea and sky would show strong continuous tilting and vertical movement. No HyperSmooth, electronic stabilization, horizon leveling, gimbal effect, handheld drift, camera orbit, tracking, panning, zooming, reframing, cuts, or reverse angles. ACTION CHECK: Head visibly airborne by 00:03. One diagonal breach, completed with a sideways impact by 00:05.5. One displacement wave, then one resurfacing. No vertical standing emergence or repeated action. The giant's gaze never interrupts their momentum. GAZE CHECK: Opening: passengers look at each other. Breach and swimming: passengers look at the giant. Final crouch: passengers avert their faces and look down or toward each other. No direct eye contact with the camera at any point. IDENTITY AND CLOTHING: A matches @image1, B matches @image2, C matches @image3. Distinct identities and exact reference garments throughout. Only wetness, natural fabric motion, and wind affect appearance. Both passengers stay wet after the wave. COLOR AND ATMOSPHERE: Cold desaturated teal-grey grading, dark rough sea, white foam, pale overcast sky, cool diffused daylight, cinematic contrast, subtle film grain. Photorealistic wet wood, skin, fabric, spray, and water. No warm sunlight or tropical colors. AUDIO: Diegetic sound only: strong wind, waves against the hull, creaking wood, rushing water, heavy sideways impact, water crossing the deck, startled gasps, frightened breathing. Opening conversation remains indistinct. No intelligible dialogue, no music. PROHIBITED: LEGO, toys, miniature people, tiny boat, narrow rowboat, cartoon rendering, whale anatomy, fins, extra passengers, duplicated faces, identity swaps, altered clothing, added swimming equipment, scale changes, vertical elevator-like emergence, hovering, slow motion, repeated breach, stationary floating head, direct camera eye contact, people falling overboard, hull destruction, obstructing mast or rigging, level stabilized sea horizon, diagonal camera direction, subtitles, captions, text overlays, added logos.

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Промпты для видео: "Понимание. Прощение. Терпение. Хватит."

2026-09-15T13:39:01.558Z

1.Создаем карту персонажа.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

Промпт для карт персонажей:

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

2.Создаем видео.
Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - Качество 480p (если мажоры, то 720р или 1080р) - Соотношение сторон любое ( у меня 4:3) - Длительность 20 сек. Загружаем КАРТУ ПЕРСОНАЖА. После вставляем промпт и вручную удаляем все "@image1", и заново их прописываем (когда будете вводить "@" там можно будет выбрать нужный файл), так они привяжутся к промпту:

ПРОМПТ ДЛЯ ВИДЕО:

Create a 20-second photorealistic cinematic drama in native 4:3 format.

CORE STORY
Show several emotionally specific moments from one person's life: teenage bullying, workplace humiliation, an abusive argument with a partner, and a painful confrontation with their mother.

The last three past incidents culminate in consecutive slaps, each in a different situation. The repeated pattern ends when the protagonist stops the next attempted slap.

This must feel like fragments of real life, not a montage of an expressionless person receiving random blows.
Every incident needs a visible emotional setup, an action, and a human reaction.

REFERENCE IDENTITY
@image1 defines PERSON_A, the protagonist.
Adapt the protagonist naturally to the gender and appearance of the reference person.

Preserve recognizable facial structure, skin tone, eyes, and hair identity.
Opening only: a recognizable teenage version, approximately 16–17 years old.
All remaining scenes: the adult protagonist.

Earlier outfits change to fit each situation.
In the FINAL scene, use the complete appearance from @image1, including the exact hairstyle, clothing, colors, and visible accessories.

The protagonist remains the SAME person in every scene.
Do not replace them with the mother, partner, or another character.

CAMERA AND BLOCKING — CRITICAL CORRECTION
Do NOT create a centered, front-facing portrait looking almost into the lens.

Use a consistent close medium shot from mid-chest upward.
Place PERSON_A slightly RIGHT of center, with room on SCREEN LEFT for the aggressor's cropped shoulder, gestures, and entering arm.
Their torso and face are turned approximately 25–35 degrees toward SCREEN LEFT, in a readable three-quarter view.
Both eyes and their emotional expressions remain visible.

The aggressor stands opposite PERSON_A, clearly to the LEFT of the camera axis, not directly behind the lens.
PERSON_A's eyeline connects with the aggressor's off-screen face.
PERSON_A NEVER looks into the camera.

Maintain approximately the same framing, face position, focal length, and three-quarter orientation across every scene.
Use a locked camera, not an orbit or a sequence of changing angles.

Show aggressors only through cropped shoulders, torsos, forearms, hands, and off-screen voices.
Keep all aggressor faces outside the frame, including reflections.
Do not hide the protagonist's face behind the aggressor's shoulder.

EDITING
Hard match cuts separate the periods of life.
Change the setting, clothing, lighting, and emotional circumstances at each cut.
No dissolves, age morphing, decorative flashback effects, or title cards.

Use natural-speed action.
The adult slaps repeat the same broad visual direction and rhythm, but each has a distinct emotional cause and reaction.
Do not use slow motion or fast-forward.

00:00–00:03 — TEENAGE BULLYING
School corridor, slightly cold daylight.
PERSON_A is a teenager in a worn casual hoodie with a backpack strap visible.

At the screen-left edge, a cropped peer's hand holds a phone displaying a small ordinary portrait of PERSON_A.
Another finger points mockingly at the photo, then toward PERSON_A.
Keep the phone secondary: the teenager's face remains the emotional focus.
Brief off-screen laughter establishes group humiliation without showing anyone else's face.

A peer shoves PERSON_A's shoulder.
They recoil slightly, clutch the backpack strap, and try to conceal their embarrassment.

Then a hand enters from BEHIND their shoulder and gives one brief staged open-handed knock to the BACK OF THE HEAD.
The contact is behind the head, never on the cheek.
Do not substitute a frontal slap.

PERSON_A lowers their eyes. A forced attempt to smile disappears.
Keep the incident non-graphic, with no injury or exaggerated head movement.

00:03–00:06 — WORKPLACE: DISMISSAL AND HUMILIATION
Hard match cut.
Adult PERSON_A wears a charcoal suit jacket and pale shirt in an office.
Preserve the three-quarter framing and screen position.

A supervisor's cropped suit sleeve enters from screen left.
The supervisor thrusts a dismissal document toward PERSON_A and angrily points toward the exit.

A sharp off-screen voice says in English:
“You’re fired!”

PERSON_A is already visibly distressed: wet eyes, trembling lower lip.
They lift an open hand in an attempt to explain and say:
“Please—”

The supervisor interrupts with one abrupt staged slap.
Make the movement decisive, not a slow touch or caress.
Use controlled cinematic staging and a synchronized dry contact sound rather than excessive force.

PERSON_A's head turns briefly with the motion.
Their unfinished hand gesture freezes, then drops.
They swallow, trying not to cry harder.

00:06–00:09 — PARTNER: AN ARGUMENT AT HOME
Hard match cut.
Evening apartment, warm practical lighting.
PERSON_A wears simple home clothing.
A male partner is represented only by a cropped torso, shoulder, and gesturing arm on screen left.

Begin in the middle of an emotionally charged argument.
PERSON_A gestures toward themselves and says, with hurt urgency:
“Listen to me!”

The partner cuts across the gesture with an accusing finger.
Their movements briefly overlap, making the conflict feel active rather than posed.
The partner then delivers one abrupt staged slap.

PERSON_A turns with the motion and falls silent.
They look back toward the partner with shock and wounded disbelief.
Tears remain visible. Their expression asks how someone close could do this.

Do not portray the protagonist's disagreement as justification for the violence.
Do not show an aggressor's face.

00:09–00:14 — MOTHER: A FAMILY CRISIS
Hard match cut to a modest kitchen under tired, uneven evening light.

PERSON_A is still the adult protagonist, not the mother.
Keep their recognizable reference face, but show a difficult period through styling:
messy unwashed hair;
creased, stained everyday clothing;
pronounced under-eye shadows;
puffy, reddened eyes;
dry lips;
slumped shoulders;
an exhausted, unhealthy appearance.

Do not turn them into a caricature or an unrecognizable different person.

At the lower edge of the frame, show an open alcohol bottle and a used drinking glass as contextual props.
Their appearance and the bottle together suggest problematic drinking; do not rely on facial appearance alone.

The MOTHER remains off-screen on the left.
Only her older hand, sleeve, and cropped shoulder are visible.
She gestures anxiously toward the bottle and reaches to move it away.
Her off-screen voice says:
“Please, stop drinking.”

PERSON_A pulls the bottle slightly back or blocks her reaching hand.
They turn toward her with tearful, defensive frustration and say:
“Mom, leave it!”

This line clearly establishes that the off-screen woman is the protagonist's mother.

The mother loses control and gives PERSON_A one brief staged slap.
The mother is the person delivering the slap.
PERSON_A is the person receiving it.
Never reverse these roles or cut to a different woman being hit.

PERSON_A goes still.
Their defensive expression collapses into hurt; tears gather and their breathing catches.
The mother's hand hesitates and withdraws, suggesting that she recognizes what she has done.

Present the slap as a harmful loss of control, not a treatment for addiction or a justified act of care.
Do not show the protagonist recovering because of the slap.

00:14–00:15.20 — FINAL CONFLICT: THE PROTAGONIST OBJECTS
Hard match cut.
PERSON_A now wears the COMPLETE outfit and appearance from @image1.
The same three-quarter framing is retained.
Use a grounded neutral interior and natural directional light, not a blank studio backdrop or heroic spotlight.

An adult male aggressor stands off-screen left, represented by a cropped shoulder and forearm.

PERSON_A is emotional, with tears in their eyes, but actively disagrees.
They raise one index finger briefly between themselves and the aggressor, without touching his face.
Their expression and gesture communicate:
“No. You are wrong.”

Their lips may move in a brief objection, but no additional intelligible line is necessary.

The aggressor begins another slap, following the repeated direction of the earlier adult incidents.

00:15.20–00:16.30 — STOP THE HAND BESIDE THE RIGHT CHEEK
The incoming open palm approaches PERSON_A's ANATOMICAL RIGHT CHEEK, not simply the right side of the image.

At the last moment, PERSON_A raises their own hand and firmly grips the aggressor's wrist.
Stop the incoming palm a few centimeters before it touches the cheek.
The halted open hand and the protagonist's face must be visible together.

The slap NEVER lands.
Do not show cheek contact followed by a delayed catch.
Do not grab the aggressor's individual fingers or create an impossible interlocked-hand pose.

Use a short, clear, anatomically plausible interception.
No wrist twisting, retaliatory punch, or wrestling.
No impact sound: the expected slap is absent.

00:16.30–00:18 — TEARS, BREATH, AND STEADY POSTURE
Hold the same shot.
PERSON_A keeps the wrist stopped briefly and looks directly at the aggressor's off-screen face.

A tear rolls down one cheek.
The protagonist is still emotionally hurt, but no longer retreats.

Show a visible controlled exhale.
Their shoulders lift slightly with the breath, draw back, then settle downward.
This is a release of tension, not a casual “whatever” shrug.
Their neck straightens, head becomes level, and their jaw relaxes.
The eyes remain wet but focused and unwavering.

Do not replace vulnerability with a smug smile or angry heroic pose.
The emotional message is:
“I can be hurt and still protect myself.”

00:18–00:19 — LOWER THE AGGRESSOR'S HAND
PERSON_A firmly guides the captured wrist downward and away from their face.
The aggressor's palm descends with the wrist in one continuous motion.
PERSON_A releases it only after the arm is outside their personal space.

No counterattack.
No shove, revenge, or pursuit.
The aggressor's hand retreats out of frame.

00:19–00:20 — FINAL EMOTIONAL HOLD
PERSON_A stays in place, shoulders settled, tears still visible.
They maintain direct eye contact with the off-screen aggressor to screen left, NEVER with the lens.

Their expression is quiet but unmistakable:
“You will not do that to me again.”

No spoken final slogan.
Let the face, breathing, and refusal to look away communicate the boundary.
End on a living expression, not a freeze-frame or fade.

PERFORMANCE PRIORITIES
Conflict comes before contact.
Every scene includes readable gestures and a specific relationship, not just an arm entering to slap someone.

The protagonist is emotionally engaged:
teenage embarrassment;
workplace desperation;
betrayal by a partner;
defensive distress with the mother;
tearful resolve in the final scene.

Use wet eyes, interrupted speech, trembling lips, small changes in breathing, and unfinished gestures.
Avoid constant theatrical sobbing, screaming in every scene, or identical reactions.

CONTACT STYLE
Slaps must read as abrupt dramatic actions, not gentle stroking.
Create intensity through the aggressor's preparation, the protagonist's reaction, editing, and restrained sound.
All contact is controlled cinematic staging.
No blood, bruises, swelling, wounds, extreme head snapping, or graphic injury.

SOUND
No music or narration.
Use brief English dialogue exactly where specified.
Voices belong to the correct characters; no overlapping unrelated lines.

School: short off-screen laughter and corridor ambience.
Office: “You’re fired!” followed by PERSON_A's interrupted “Please—”.
Partner: PERSON_A says “Listen to me!”
Mother: “Please, stop drinking.” PERSON_A replies, “Mom, leave it!”

Use restrained contact sounds for the three adult slaps.
In the finale, replace the expected slap sound with fabric movement and a controlled exhale.
Lower the ambient sound slightly during the final hold.

AVOID
Centered frontal portrait posing;
looking into the camera;
aggressors positioned directly behind the lens;
visible aggressor faces;
random slaps without emotional setup;
gentle cheek stroking instead of staged slaps;
a cheek slap replacing the teenage back-of-head incident;
the mother becoming the person who is hit;
a different woman replacing the protagonist;
an immaculate appearance during the drinking-crisis scene;
portraying violence as treatment or deserved punishment;
a final slap landing before the interception;
wrong-side cheek contact;
extra hands, fused fingers, broken wrists, identity changes;
revenge violence;
text, captions, subtitles, watermarks, black bars.

Как сделать видео, где вы битой разбиваете бутылку?

2026-09-15T09:12:18.543Z

*В конце будет 3 промпта, (детский, женский и мужской

1.Создаем карту персонажа.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

Промпт для карт персонажей:

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

RIGHT PANEL (close-up):

A close-up portrait of the person's face, cropped at the shoulders (bust shot). Face front-facing, looking directly at the camera, neutral expression, natural even lighting. Preserve all facial features and details as accurately as possible: eye shape, nose, lips, eyebrows, face shape, eye color, skin tone and texture, hair shape and color, as well as any unique features (moles, freckles, ear shape). High detail, realism, portrait quality.

LEFT PANEL (back view, full body):

The person standing at full height, front view, in their everyday clothing, arms along the body, feet shoulder-width apart. IMPORTANT: the head and face are completely absent — the figure is cropped at the neck/shoulder line, no head. Only the figure from the neck down is shown. Preserve body build, proportions, and clothing.

MIDDLE PANEL (front view, full body):

The same person, back view, in the same clothing, at full height, including the head and hair (viewed from behind). Preserve the hairstyle, body build, and clothing details.

GENERAL REQUIREMENTS:

Unified style, consistent lighting and figure scale across all panels, consistency of appearance and clothing between panels. Clean background, clear vertic

2.Создаем видео.
Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - Качество 480p (если мажоры, то 720р или 1080р) - Соотношение сторон 9:16 - Длительность 10 сек. Загружаем КАРТУ ПЕРСОНАЖА. После вставляем промпт и вручную удаляем все "@image1", и заново их прописываем (когда будете вводить "@" там можно будет выбрать нужный файл), так они привяжутся к промпту:

- Промпт с шампанским:

@image1 defines the main character. Preserve their complete appearance: gender, age, face, hairstyle, skin tone, body proportions, exact clothing, makeup (if any), and all visible details. Do not change or redesign anything from @image1. A 10-second vertical 9:16 cinematic short film shot in one continuous take with a fixed camera position. A person walks into frame, sits on the EDGE of a simple wooden bench, tries to open a champagne bottle by hitting the BOTTOM of the bottle against the bench edge multiple times, fails, catches a baseball bat thrown from off-screen, and smashes the bottle in an epic slow-motion explosion of glass, foam, and liquid. Autumn setting with golden-hour lighting and beautiful fall foliage. Professional commercial-quality cinematography. SETTING AND COMPOSITION Location: Outdoor autumn park or backyard setting. A simple WOODEN BENCH (NOT a table, NOT a picnic table) is positioned in the LOWER CENTER of the frame, running horizontally across the bottom third. The bench is a single flat horizontal wooden plank (light natural wood color) supported by simple legs or a frame — the kind of park bench or garden bench you sit on. The bench must be clearly visible, centered at the bottom of the frame, NOT shifted to the left or right. Behind the bench: dense beautiful autumn trees and bushes with thick, vibrant yellow, orange, and red leaves — the foliage should look lush, rich, and visually stunning, creating a warm autumn atmosphere. Green grass lawn visible on the ground. Lighting: Golden hour — warm late afternoon sunlight. Strong backlight from behind the trees creates a beautiful rim light and lens flare around the person. Additional soft fill light from the front illuminates the person's face naturally and professionally. The entire scene has warm, cinematic autumn tones. Camera: Fixed wide shot, 9:16 vertical format. The camera does not move, zoom, pan, or cut throughout the entire 10 seconds. The person's face stays centered in the upper-middle portion of the frame throughout. The bench remains visible and centered in the lower third of the frame. Frame the shot so there is enough headroom above and enough space below to see the bench clearly and the exploding debris at the end. THE CHARACTER (from @image1) The main character is the person from @image1. Use their exact gender, age, face, hairstyle, clothing, and overall appearance without any changes. They enter the frame confidently, perform the actions described below, and maintain a cool, slightly defiant attitude throughout — especially during the final bat swing and explosion. CRITICAL: The person does NOT speak, does NOT move their lips, does NOT say anything at any point in the video. They remain silent throughout. THE BOTTLE A champagne bottle (standard 750ml size, green glass, gold or silver foil wrapper around the neck and top). CRITICAL: The bottle is SEALED and UNOPENED. At the top of the bottle there is a champagne cork (mushroom-shaped cork stopper) with a wire cage (muselet) twisted around it to hold the cork in place — this is the standard champagne closure. The cork is clearly visible protruding slightly from the bottle neck, covered by the foil at the very top. The bottle has NOT been opened — the cork is still in place, the wire cage is still secured. The bottle looks like an unopened, full champagne bottle ready to be opened, NOT an already-opened bottle with an empty neck. The person holds the bottle by the NECK (upper part, below the cork), NOT by the bottom. THE BAT A wooden baseball bat (natural wood grain, classic baseball bat shape and size). It is thrown into frame from OFF-SCREEN LEFT and caught by the person. -— TIMELINE — 10 SECONDS, ONE CONTINUOUS SHOT, NO CUTS [00:00 – 00:02] ENTRANCE The person walks into the frame from the LEFT side, moving naturally and confidently toward the bench. They hold the champagne bottle upright by the neck in one hand (cork and wire cage clearly visible at the top). The camera does not follow them — they simply walk into the stationary frame. Their face is visible and centered as they approach the bench. Calm, focused expression. The autumn trees and golden backlight are visible throughout. Audio: footsteps on grass, ambient autumn sounds (light breeze, distant birds), natural outdoor atmosphere. [00:02 – 00:03] POSITIONING AND SITTING CRITICAL MOMENT — THE PERSON SITS DOWN IN A SPECIFIC WAY: The person reaches the bench and positions themselves at the FAR SIDE of the bench (the side AWAY from the camera — the outer edge). They turn slightly sideways (three-quarter angle) so their face and upper body remain clearly visible to the camera. They lower themselves and SIT DOWN on the EDGE of the bench — their body weight is supported by the bench, they are seated, NOT standing. They sit in a casual, slightly leaning-forward posture with one leg forward for balance. The bench is clearly visible beneath them. They are sitting ON the bench, not standing beside it. They hold the sealed champagne bottle (with visible cork and wire cage at the top) in one hand by the neck, and they look down at the bench edge in front of them, preparing to strike. Audio: quiet movement, fabric rustling, the slight creak of the bench as they sit. [00:03 – 00:04] FIRST HIT — BOTTOM OF BOTTLE AGAINST BENCH EDGE CRITICAL: The person holds the champagne bottle by the NECK (upper part, below the cork and wire cage). They raise the bottle and swing it downward so the BOTTOM (base) of the bottle strikes the EDGE of the wooden bench — a sharp, deliberate strike. The BOTTOM hits the EDGE, NOT the top surface. The cork stays in place — the bottle does NOT break or open. Nothing happens. The person holds the bottle still for a brief beat, staring at it with a slightly confused or annoyed expression. THEIR MOUTH STAYS CLOSED — no speaking, no lip movement. Audio: a sharp THUD of glass hitting wood, no breaking sound. [00:04 – 00:05] LOOK AT CAMERA — "WHAT?" (SILENT) The person lifts their gaze from the bottle and looks DIRECTLY INTO THE CAMERA with a brief, intense, slightly threatening expression — eyebrows slightly furrowed, a "what the hell?" or "are you kidding me?" look. THEIR MOUTH STAYS CLOSED. No speaking. No words. Just a silent, intense stare. The stare lasts about half a second, then they look back down at the bottle. Audio: silence, tension, perhaps a quiet exhale or scoff, but NO SPEECH. [00:05 – 00:06.5] REPEATED HITS — BOTTOM AGAINST EDGE The person strikes the BOTTOM of the bottle against the EDGE of the bench two or three more times in quick succession — THUD, THUD, THUD — each hit firm and frustrated. The cork stays in place — the bottle still does NOT break or open. The person's expression grows more annoyed and determined. THEIR MOUTH STAYS CLOSED throughout — no speaking, no lip movement. They keep their body positioned so their face stays visible and centered. Audio: repeated THUD sounds, slightly louder and more forceful each time, the person's breathing slightly heavier, but NO SPEECH OR DIALOGUE. [00:06.5 – 00:07] ANOTHER LOOK AT CAMERA, THEN DECISION (SILENT) The person lifts their gaze to the camera again — this time a longer, more deliberate stare, full of quiet confidence and defiance. A "fine, I'll do it my way" look. THEIR MOUTH STAYS CLOSED. No speaking. They straighten up slightly, set the champagne bottle down UPRIGHT on the bench surface (cork and wire cage pointing up, bottle standing stable), and stand up or shift their weight forward. Their hands are now free. They look off to the LEFT, ready. Audio: the bottle is set down with a light CLINK on wood, ambient sound, anticipation builds, but NO SPEECH. [00:07 – 00:08] THE BAT IS THROWN AND CAUGHT A wooden baseball bat is thrown INTO THE FRAME FROM OFF-SCREEN LEFT (from outside the left edge of the frame). It travels horizontally through the air toward the person. The person reacts instantly, reaches out with both hands, and CATCHES the bat firmly — one hand grabs the handle, the other steadies the barrel. The catch is clean and confident. They immediately adjust their grip and prepare to swing. Their expression is focused, determined, and ready. THEIR MOUTH STAYS CLOSED. Audio: the BAT WHOOSHES through the air, a solid SMACK as it lands in their hands, a brief moment of silence as they grip it. [00:08 – 00:08.5] THE SWING BEGINS (REAL-TIME) The person steps slightly forward (or remains seated but leans forward), raises the bat with both hands above and behind their shoulder (classic baseball swing wind-up), and begins the downward swing toward the sealed champagne bottle (with cork visible) standing on the bench. As the bat descends, their eyes LIFT FROM THE BOTTLE AND LOCK ONTO THE CAMERA — the gaze shifts upward the moment the bat is in motion, before contact is made. The expression is cool, confident, defiant, powerful — "watch this." THEIR MOUTH STAYS CLOSED. Audio: the WHOOSH of the bat cutting through the air, building intensity. [00:08.5 – 00:10] THE IMPACT AND EXPLOSION (SLOW MOTION) The moment the bat makes contact with the champagne bottle, TIME SLOWS DOWN to dramatic slow motion (roughly 40–50% speed). This slow-motion section lasts for the final 1.5–2 seconds of the video. The bat STRIKES the bottle. The bottle EXPLODES in an epic, highly detailed burst of: - Glass shards and fragments flying outward in all directions - Champagne liquid spraying and splashing in streams and droplets - White foam erupting and expanding outward - Cork, wire cage, and foil pieces spinning through the air - Small sparkling droplets catching the golden light The debris explodes in ALL DIRECTIONS: - Some pieces fly to the RIGHT - Some pieces fly to the LEFT - Some pieces fly DIRECTLY TOWARD THE CAMERA (coming at the lens, growing larger as they approach) - Some pieces fly upward and downward The person's face remains CENTERED and CLEARLY VISIBLE throughout the explosion. Their eyes are locked on the camera the entire time, expression steady, confident, and slightly smirking or serious — a "that's how it's done" look. THEIR MOUTH STAYS CLOSED — no speaking, no shouting, no lip movement. The bat is frozen mid-follow-through. The golden backlight catches the flying droplets and glass, making them sparkle and glow beautifully. The autumn trees and warm light remain visible in the background. The explosion expands and fills MORE AND MORE OF THE FRAME as the slow motion continues. By the final moment, the debris cloud occupies a LARGE PORTION of the screen — glass, foam, liquid particles, cork pieces are everywhere, suspended in the air, beautifully lit, with the person's confident face still centered and visible behind/within the cloud. Audio: the CRACK of the impact is deep and resonant in slow motion, followed by the slow, bassy ROAR of the explosion — shattering glass, rushing liquid, foam expanding, and the cork shooting through the air, all stretched out and emphasized. Cinematic, powerful, satisfying sound design. NO SPEECH, NO DIALOGUE. [00:10] END The video ENDS on this final slow-motion frame: the debris at maximum expansion, filling much of the screen, the person's face centered and staring confidently into the camera, frozen in the aftermath of the explosion. No fade, no cut to black — just hold this epic moment and end. -— GLOBAL CONSTRAINTS AND PRESERVATION Camera: FIXED POSITION throughout the entire 10 seconds. No movement, no zoom, no pan, no handheld shake. The framing stays locked. 9:16 vertical format. Continuity: This is ONE CONTINUOUS SHOT with no cuts, no scene changes, no jump cuts. The only speed change is the transition to slow motion at [00:08.5]. The person's face: stays CENTERED in the frame throughout the entire video. Always clearly visible, always in focus. The bench: A simple wooden bench (NOT a table), centered in the lower third of the frame, running horizontally. The person SITS ON THE EDGE of this bench (not standing beside it) when hitting the bottle. The bottle: champagne, not Coca-Cola. Green glass, gold or silver foil wrapper. SEALED with a visible mushroom-shaped cork and wire cage (muselet) at the top. The bottle is UNOPENED — the cork is in place throughout until the bottle is smashed by the bat. The person holds it by the NECK and hits the BOTTOM against the bench EDGE. The bat: wooden, natural grain, classic baseball bat. Autumn atmosphere: warm golden hour light, backlight with lens flare, dense yellow/orange/red fall foliage in the background, green grass, beautiful autumn tones throughout. Lighting: professional cinematic lighting — strong backlight (rim light, lens flare), soft frontal fill light on the person's face. Clothing and appearance: taken 100% from @image1 without changes. Audio: diegetic sound only — footsteps, movement, bottle hitting bench (THUD), bat whooshing and catching, the swing, the impact (CRACK), the explosion (slow-motion shattering and liquid spray). NO SPEECH. NO DIALOGUE. NO WORDS SPOKEN. NO LIP MOVEMENT. No background music, no subtitles. NEGATIVE CONSTRAINTS No cuts or jump cuts (except the speed ramp to slow motion at the end). No camera movement or zoom. No background music or soundtrack. No subtitles or text overlay. NO SPEECH. NO DIALOGUE. NO WORDS. The person does NOT speak, does NOT move their lips, does NOT say anything at any point in the video. Complete silence from the person. Do not change the person's appearance, clothing, gender, or age from @image1. Do not use a Coca-Cola bottle — use champagne. The champagne bottle must be SEALED with a visible cork and wire cage (muselet) at the top. Do not show an already-opened bottle or a bottle with an empty neck. The cork must be clearly visible protruding from the bottle neck. Do not use a metal bat — use wood. Do not use a TABLE — use a simple wooden BENCH, centered at the bottom of the frame. The person SITS ON THE EDGE of the bench when hitting the bottle — they do NOT stand. The person hits the BOTTOM of the bottle against the EDGE of the bench — NOT the top/neck of the bottle, and NOT against a flat table surface. The bottle does NOT open from the bench hits — it only breaks when hit with the bat. Do not show another person throwing the bat — the bat simply flies in from off-screen left.

- Промпт с пивом:

@image1 defines the main character. Preserve their complete appearance: gender, age, face, hairstyle, skin tone, body proportions, exact clothing, makeup (if any), and all visible details. Do not change or redesign anything from @image1. A 10-second vertical 9:16 cinematic short film shot in one continuous take with a fixed camera position. A person walks into frame, sits on the EDGE of a simple wooden bench, tries to open a glass beer bottle by hitting the BOTTOM of the bottle against the bench edge multiple times, fails, catches a baseball bat thrown from off-screen, and smashes the bottle in an epic slow-motion explosion of glass, beer foam, and liquid. Autumn setting with golden-hour lighting and beautiful fall foliage. Professional commercial-quality cinematography. SETTING AND COMPOSITION Location: Outdoor autumn park or backyard setting. A simple WOODEN BENCH (NOT a table, NOT a picnic table) is positioned in the LOWER CENTER of the frame, running horizontally across the bottom third. The bench is a single flat horizontal wooden plank (light natural wood color) supported by simple legs or a frame — the kind of park bench or garden bench you sit on. The bench must be clearly visible, centered at the bottom of the frame, NOT shifted to the left or right. Behind the bench: dense beautiful autumn trees and bushes with thick, vibrant yellow, orange, and red leaves — the foliage should look lush, rich, and visually stunning, creating a warm autumn atmosphere. Green grass lawn visible on the ground. Lighting: Golden hour — warm late afternoon sunlight. Strong backlight from behind the trees creates a beautiful rim light and lens flare around the person. Additional soft fill light from the front illuminates the person's face naturally and professionally. The entire scene has warm, cinematic autumn tones. Camera: Fixed wide shot, 9:16 vertical format. The camera does not move, zoom, pan, or cut throughout the entire 10 seconds. The person's face stays centered in the upper-middle portion of the frame throughout. The bench remains visible and centered in the lower third of the frame. Frame the shot so there is enough headroom above and enough space below to see the bench clearly and the exploding debris at the end. THE CHARACTER (from @image1) The main character is the person from @image1. Use their exact gender, age, face, hairstyle, clothing, and overall appearance without any changes. They enter the frame confidently, perform the actions described below, and maintain a cool, slightly defiant attitude throughout — especially during the final bat swing and explosion. CRITICAL: The person does NOT speak, does NOT move their lips, does NOT say anything at any point in the video. They remain silent throughout. THE BOTTLE A glass beer bottle (standard 330ml or 500ml size, dark brown or green glass). CRITICAL: The bottle is SEALED and UNOPENED. At the top of the bottle there is a metal bottle cap (crown cap) crimped firmly onto the bottle neck — this is the standard beer bottle closure. The cap is clearly visible at the top of the bottle. The bottle has a beer label wrapped around the middle. The bottle has NOT been opened — the metal cap is still firmly in place, completely sealed. The bottle looks like an unopened, full glass beer bottle, NOT an already-opened bottle or a bottle without a cap. The person holds the bottle by the NECK (upper part, below the cap), NOT by the bottom. THE BAT A wooden baseball bat (natural wood grain, classic baseball bat shape and size). It is thrown into frame from OFF-SCREEN LEFT and caught by the person. -— TIMELINE — 10 SECONDS, ONE CONTINUOUS SHOT, NO CUTS [00:00 – 00:02] ENTRANCE The person walks into the frame from the LEFT side, moving naturally and confidently toward the bench. They hold the glass beer bottle upright by the neck in one hand (metal cap clearly visible at the top, beer label visible on the body). The camera does not follow them — they simply walk into the stationary frame. Their face is visible and centered as they approach the bench. Calm, focused expression. The autumn trees and golden backlight are visible throughout. Audio: footsteps on grass, ambient autumn sounds (light breeze, distant birds), natural outdoor atmosphere. [00:02 – 00:03] POSITIONING AND SITTING CRITICAL MOMENT — THE PERSON SITS DOWN IN A SPECIFIC WAY: The person reaches the bench and positions themselves at the FAR SIDE of the bench (the side AWAY from the camera — the outer edge). They turn slightly sideways (three-quarter angle) so their face and upper body remain clearly visible to the camera. They lower themselves and SIT DOWN on the EDGE of the bench — their body weight is supported by the bench, they are seated, NOT standing. They sit in a casual, slightly leaning-forward posture with one leg forward for balance. The bench is clearly visible beneath them. They are sitting ON the bench, not standing beside it. They hold the sealed glass beer bottle (with visible metal cap at the top and label on the body) in one hand by the neck, and they look down at the bench edge in front of them, preparing to strike. Audio: quiet movement, fabric rustling, the slight creak of the bench as they sit. [00:03 – 00:04] FIRST HIT — BOTTOM OF BOTTLE AGAINST BENCH EDGE CRITICAL: The person holds the glass beer bottle by the NECK (upper part, below the metal cap). They raise the bottle and swing it downward so the BOTTOM (base) of the bottle strikes the EDGE of the wooden bench — a sharp, deliberate strike. The BOTTOM hits the EDGE, NOT the top surface. The metal cap stays in place — the bottle does NOT break or open. Nothing happens. The person holds the bottle still for a brief beat, staring at it with a slightly confused or annoyed expression. THEIR MOUTH STAYS CLOSED — no speaking, no lip movement. Audio: a sharp THUD of glass hitting wood, no breaking sound. [00:04 – 00:05] LOOK AT CAMERA — "WHAT?" (SILENT) The person lifts their gaze from the bottle and looks DIRECTLY INTO THE CAMERA with a brief, intense, slightly threatening expression — eyebrows slightly furrowed, a "what the hell?" or "are you kidding me?" look. THEIR MOUTH STAYS CLOSED. No speaking. No words. Just a silent, intense stare. The stare lasts about half a second, then they look back down at the bottle. Audio: silence, tension, perhaps a quiet exhale or scoff, but NO SPEECH. [00:05 – 00:06.5] REPEATED HITS — BOTTOM AGAINST EDGE The person strikes the BOTTOM of the glass beer bottle against the EDGE of the bench two or three more times in quick succession — THUD, THUD, THUD — each hit firm and frustrated. The metal cap stays in place — the bottle still does NOT break or open. The person's expression grows more annoyed and determined. THEIR MOUTH STAYS CLOSED throughout — no speaking, no lip movement. They keep their body positioned so their face stays visible and centered. Audio: repeated THUD sounds, slightly louder and more forceful each time, the person's breathing slightly heavier, but NO SPEECH OR DIALOGUE. [00:06.5 – 00:07] ANOTHER LOOK AT CAMERA, THEN DECISION (SILENT) The person lifts their gaze to the camera again — this time a longer, more deliberate stare, full of quiet confidence and defiance. A "fine, I'll do it my way" look. THEIR MOUTH STAYS CLOSED. No speaking. They straighten up slightly, set the glass beer bottle down UPRIGHT on the bench surface (metal cap pointing up, label visible, bottle standing stable), and stand up or shift their weight forward. Their hands are now free. They look off to the LEFT, ready. Audio: the bottle is set down with a light CLINK on wood, ambient sound, anticipation builds, but NO SPEECH. [00:07 – 00:08] THE BAT IS THROWN AND CAUGHT A wooden baseball bat is thrown INTO THE FRAME FROM OFF-SCREEN LEFT (from outside the left edge of the frame). It travels horizontally through the air toward the person. The person reacts instantly, reaches out with both hands, and CATCHES the bat firmly — one hand grabs the handle, the other steadies the barrel. The catch is clean and confident. They immediately adjust their grip and prepare to swing. Their expression is focused, determined, and ready. THEIR MOUTH STAYS CLOSED. Audio: the BAT WHOOSHES through the air, a solid SMACK as it lands in their hands, a brief moment of silence as they grip it. [00:08 – 00:08.5] THE SWING BEGINS (REAL-TIME) The person steps slightly forward (or remains seated but leans forward), raises the bat with both hands above and behind their shoulder (classic baseball swing wind-up), and begins the downward swing toward the sealed glass beer bottle (with metal cap visible and label visible) standing on the bench. As the bat descends, their eyes LIFT FROM THE BOTTLE AND LOCK ONTO THE CAMERA — the gaze shifts upward the moment the bat is in motion, before contact is made. The expression is cool, confident, defiant, powerful — "watch this." THEIR MOUTH STAYS CLOSED. Audio: the WHOOSH of the bat cutting through the air, building intensity. [00:08.5 – 00:10] THE IMPACT AND EXPLOSION (SLOW MOTION) The moment the bat makes contact with the glass beer bottle, TIME SLOWS DOWN to dramatic slow motion (roughly 40–50% speed). This slow-motion section lasts for the final 1.5–2 seconds of the video. The bat STRIKES the bottle. The bottle EXPLODES in an epic, highly detailed burst of: - Glass shards and fragments flying outward in all directions - Beer liquid spraying and splashing in golden-amber streams and droplets - White beer foam erupting and expanding outward - Metal bottle cap spinning and tumbling through the air - Small sparkling droplets catching the golden hour light beautifully The debris explodes in ALL DIRECTIONS: - Some pieces fly to the RIGHT - Some pieces fly to the LEFT - Some pieces fly DIRECTLY TOWARD THE CAMERA (coming at the lens, growing larger as they approach) - Some pieces fly upward and downward The person's face remains CENTERED and CLEARLY VISIBLE throughout the explosion. Their eyes are locked on the camera the entire time, expression steady, confident, and slightly smirking or serious — a "that's how it's done" look. THEIR MOUTH STAYS CLOSED — no speaking, no shouting, no lip movement. The bat is frozen mid-follow-through. The golden backlight catches the flying amber beer droplets and glass shards, making them glow and sparkle beautifully against the autumn background. The autumn trees and warm light remain visible in the background. The explosion expands and fills MORE AND MORE OF THE FRAME as the slow motion continues. By the final moment, the debris cloud occupies a LARGE PORTION of the screen — glass shards, golden beer foam, amber liquid droplets, and the metal cap are everywhere, suspended in the air, beautifully lit, with the person's confident face still centered and visible behind/within the cloud. Audio: the CRACK of the impact is deep and resonant in slow motion, followed by the slow, bassy ROAR of the explosion — shattering glass, rushing beer liquid, foam expanding, and the cap shooting through the air, all stretched out and emphasized. Cinematic, powerful, satisfying sound design. NO SPEECH, NO DIALOGUE. [00:10] END The video ENDS on this final slow-motion frame: the debris at maximum expansion, filling much of the screen, the person's face centered and staring confidently into the camera, frozen in the aftermath of the explosion. No fade, no cut to black — just hold this epic moment and end. -— GLOBAL CONSTRAINTS AND PRESERVATION Camera: FIXED POSITION throughout the entire 10 seconds. No movement, no zoom, no pan, no handheld shake. The framing stays locked. 9:16 vertical format. Continuity: This is ONE CONTINUOUS SHOT with no cuts, no scene changes, no jump cuts. The only speed change is the transition to slow motion at [00:08.5]. The person's face: stays CENTERED in the frame throughout the entire video. Always clearly visible, always in focus. The bench: A simple wooden bench (NOT a table), centered in the lower third of the frame, running horizontally. The person SITS ON THE EDGE of this bench (not standing beside it) when hitting the bottle. The bottle: glass beer bottle (NOT plastic, NOT champagne, NOT Coca-Cola). Dark brown or green glass with amber/dark liquid inside and a beer label on the body. SEALED with a visible metal bottle cap (crown cap) at the top. The bottle is UNOPENED — the metal cap is in place throughout until the bottle is smashed by the bat. The person holds it by the NECK and hits the BOTTOM against the bench EDGE. The bat: wooden, natural grain, classic baseball bat. Autumn atmosphere: warm golden hour light, backlight with lens flare, dense yellow/orange/red fall foliage in the background, green grass, beautiful autumn tones throughout. Lighting: professional cinematic lighting — strong backlight (rim light, lens flare), soft frontal fill light on the person's face. Clothing and appearance: taken 100% from @image1 without changes. Audio: diegetic sound only — footsteps, movement, bottle hitting bench (THUD), bat whooshing and catching, the swing, the impact (CRACK), the explosion (slow-motion shattering and liquid spray). NO SPEECH. NO DIALOGUE. NO WORDS SPOKEN. NO LIP MOVEMENT. No background music, no subtitles. NEGATIVE CONSTRAINTS No cuts or jump cuts (except the speed ramp to slow motion at the end). No camera movement or zoom. No background music or soundtrack. No subtitles or text overlay. NO SPEECH. NO DIALOGUE. NO WORDS. The person does NOT speak, does NOT move their lips, does NOT say anything at any point in the video. Complete silence from the person. Do not change the person's appearance, clothing, gender, or age from @image1. Do not use a champagne bottle, a Coca-Cola bottle, or any other bottle — use a glass beer bottle with a metal crown cap and a beer label. Do not use a plastic bottle — use GLASS only. The beer bottle must be SEALED with a visible metal bottle cap (crown cap) at the top. Do not show an already-opened bottle or a bottle without a cap. The cap must be clearly visible. Do not use a metal bat — use wood. Do not use a TABLE — use a simple wooden BENCH, centered at the bottom of the frame. The person SITS ON THE EDGE of the bench when hitting the bottle — they do NOT stand. The person hits the BOTTOM of the bottle against the EDGE of the bench — NOT the top/neck of the bottle, and NOT against a flat table surface. The bottle does NOT open from the bench hits — it only breaks when hit with the bat. Do not show another person throwing the bat — the bat simply flies in from off-screen left.

- Промпт с колой:

@image1 defines the main character. Preserve their complete appearance: gender, age, face, hairstyle, skin tone, body proportions, exact clothing, and all visible details. Do not change or redesign anything from @image1. This is a child or teenager — keep their youthful appearance exactly as shown in the reference. A 10-second vertical 9:16 cinematic short film shot in one continuous take with a fixed camera position. A child or teenager walks into frame, sits on the EDGE of a simple wooden bench, tries to open a glass Coca-Cola bottle by hitting the BOTTOM of the bottle against the bench edge multiple times, fails, catches a baseball bat thrown from off-screen, and smashes the bottle in an epic slow-motion explosion of glass, soda, and foam. Autumn setting with golden-hour lighting and beautiful fall foliage. Professional commercial-quality cinematography. SETTING AND COMPOSITION Location: Outdoor autumn park or backyard setting. A simple WOODEN BENCH (NOT a table, NOT a picnic table) is positioned in the LOWER CENTER of the frame, running horizontally across the bottom third. The bench is a single flat horizontal wooden plank (light natural wood color) supported by simple legs or a frame — the kind of park bench or garden bench you sit on. The bench must be clearly visible, centered at the bottom of the frame, NOT shifted to the left or right. Behind the bench: dense beautiful autumn trees and bushes with thick, vibrant yellow, orange, and red leaves — the foliage should look lush, rich, and visually stunning, creating a warm autumn atmosphere. Green grass lawn visible on the ground. Lighting: Golden hour — warm late afternoon sunlight. Strong backlight from behind the trees creates a beautiful rim light and lens flare around the person. Additional soft fill light from the front illuminates the person's face naturally and professionally. The entire scene has warm, cinematic autumn tones. Camera: Fixed wide shot, 9:16 vertical format. The camera does not move, zoom, pan, or cut throughout the entire 10 seconds. The person's face stays centered in the upper-middle portion of the frame throughout. The bench remains visible and centered in the lower third of the frame. Frame the shot so there is enough headroom above and enough space below to see the bench clearly and the exploding debris at the end. THE CHARACTER (from @image1) The main character is the child or teenager from @image1. Use their exact gender, age, face, hairstyle, clothing, and overall appearance without any changes. If the reference shows a boy, the character is a boy. If the reference shows a girl, the character is a girl. Keep their youthful appearance, energy, and proportions exactly as shown. They enter the frame confidently, perform the actions described below, and maintain a cool, slightly defiant attitude throughout — especially during the final bat swing and explosion. CRITICAL: The person does NOT speak, does NOT move their lips, does NOT say anything at any point in the video. They remain silent throughout. THE BOTTLE A glass Coca-Cola bottle (classic contoured glass bottle shape, approximately 330ml or 500ml size, clear glass with dark brown/black soda visible inside). CRITICAL: The bottle is SEALED and UNOPENED. At the top of the bottle there is a red metal bottle cap (crown cap) crimped onto the bottle neck — this is the standard Coca-Cola closure. The cap is clearly visible at the top. The bottle has the iconic red and white Coca-Cola label wrapped around the middle. The bottle has NOT been opened — the red cap is still in place, firmly sealed. The bottle looks like an unopened, full glass Coca-Cola bottle, NOT an already-opened bottle or a plastic bottle. The person holds the bottle by the NECK (upper part, below the cap), NOT by the bottom. THE BAT A wooden baseball bat (natural wood grain, classic baseball bat shape and size). It is thrown into frame from OFF-SCREEN LEFT and caught by the person. -— TIMELINE — 10 SECONDS, ONE CONTINUOUS SHOT, NO CUTS [00:00 – 00:02] ENTRANCE The child or teenager walks into the frame from the LEFT side, moving naturally and confidently toward the bench. They hold the glass Coca-Cola bottle upright by the neck in one hand (red cap clearly visible at the top, Coca-Cola label visible). The camera does not follow them — they simply walk into the stationary frame. Their face is visible and centered as they approach the bench. Calm, focused expression. The autumn trees and golden backlight are visible throughout. Audio: footsteps on grass, ambient autumn sounds (light breeze, distant birds), natural outdoor atmosphere. [00:02 – 00:03] POSITIONING AND SITTING CRITICAL MOMENT — THE PERSON SITS DOWN IN A SPECIFIC WAY: The child or teenager reaches the bench and positions themselves at the FAR SIDE of the bench (the side AWAY from the camera — the outer edge). They turn slightly sideways (three-quarter angle) so their face and upper body remain clearly visible to the camera. They lower themselves and SIT DOWN on the EDGE of the bench — their body weight is supported by the bench, they are seated, NOT standing. They sit in a casual, slightly leaning-forward posture with one leg forward for balance. The bench is clearly visible beneath them. They are sitting ON the bench, not standing beside it. They hold the sealed glass Coca-Cola bottle (with visible red cap at the top and Coca-Cola label) in one hand by the neck, and they look down at the bench edge in front of them, preparing to strike. Audio: quiet movement, fabric rustling, the slight creak of the bench as they sit. [00:03 – 00:04] FIRST HIT — BOTTOM OF BOTTLE AGAINST BENCH EDGE CRITICAL: The person holds the glass Coca-Cola bottle by the NECK (upper part, below the red cap). They raise the bottle and swing it downward so the BOTTOM (base) of the bottle strikes the EDGE of the wooden bench — a sharp, deliberate strike. The BOTTOM hits the EDGE, NOT the top surface. The red cap stays in place — the bottle does NOT break or open. Nothing happens. The person holds the bottle still for a brief beat, staring at it with a slightly confused or annoyed expression. THEIR MOUTH STAYS CLOSED — no speaking, no lip movement. Audio: a sharp THUD of glass hitting wood, no breaking sound. [00:04 – 00:05] LOOK AT CAMERA — "WHAT?" (SILENT) The person lifts their gaze from the bottle and looks DIRECTLY INTO THE CAMERA with a brief, intense, slightly threatening expression — eyebrows slightly furrowed, a "what the hell?" or "are you kidding me?" look. THEIR MOUTH STAYS CLOSED. No speaking. No words. Just a silent, intense stare. The stare lasts about half a second, then they look back down at the bottle. Audio: silence, tension, perhaps a quiet exhale or scoff, but NO SPEECH. [00:05 – 00:06.5] REPEATED HITS — BOTTOM AGAINST EDGE The person strikes the BOTTOM of the glass Coca-Cola bottle against the EDGE of the bench two or three more times in quick succession — THUD, THUD, THUD — each hit firm and frustrated. The red cap stays in place — the bottle still does NOT break or open. The person's expression grows more annoyed and determined. THEIR MOUTH STAYS CLOSED throughout — no speaking, no lip movement. They keep their body positioned so their face stays visible and centered. Audio: repeated THUD sounds, slightly louder and more forceful each time, the person's breathing slightly heavier, but NO SPEECH OR DIALOGUE. [00:06.5 – 00:07] ANOTHER LOOK AT CAMERA, THEN DECISION (SILENT) The person lifts their gaze to the camera again — this time a longer, more deliberate stare, full of quiet confidence and defiance. A "fine, I'll do it my way" look. THEIR MOUTH STAYS CLOSED. No speaking. They straighten up slightly, set the glass Coca-Cola bottle down UPRIGHT on the bench surface (red cap pointing up, Coca-Cola label visible, bottle standing stable), and stand up or shift their weight forward. Their hands are now free. They look off to the LEFT, ready. Audio: the bottle is set down with a light CLINK on wood, ambient sound, anticipation builds, but NO SPEECH. [00:07 – 00:08] THE BAT IS THROWN AND CAUGHT A wooden baseball bat is thrown INTO THE FRAME FROM OFF-SCREEN LEFT (from outside the left edge of the frame). It travels horizontally through the air toward the person. The person reacts instantly, reaches out with both hands, and CATCHES the bat firmly — one hand grabs the handle, the other steadies the barrel. The catch is clean and confident. They immediately adjust their grip and prepare to swing. Their expression is focused, determined, and ready. THEIR MOUTH STAYS CLOSED. Audio: the BAT WHOOSHES through the air, a solid SMACK as it lands in their hands, a brief moment of silence as they grip it. [00:08 – 00:08.5] THE SWING BEGINS (REAL-TIME) The person steps slightly forward (or remains seated but leans forward), raises the bat with both hands above and behind their shoulder (classic baseball swing wind-up), and begins the downward swing toward the sealed glass Coca-Cola bottle (with red cap visible, Coca-Cola label visible) standing on the bench. As the bat descends, their eyes LIFT FROM THE BOTTLE AND LOCK ONTO THE CAMERA — the gaze shifts upward the moment the bat is in motion, before contact is made. The expression is cool, confident, defiant, powerful — "watch this." THEIR MOUTH STAYS CLOSED. Audio: the WHOOSH of the bat cutting through the air, building intensity. [00:08.5 – 00:10] THE IMPACT AND EXPLOSION (SLOW MOTION) The moment the bat makes contact with the glass Coca-Cola bottle, TIME SLOWS DOWN to dramatic slow motion (roughly 40–50% speed). This slow-motion section lasts for the final 1.5–2 seconds of the video. The bat STRIKES the bottle. The bottle EXPLODES in an epic, highly detailed burst of: - Glass shards and fragments flying outward in all directions - Dark brown Coca-Cola liquid spraying and splashing in streams and droplets - Foam and carbonation erupting and expanding outward - Red bottle cap spinning through the air - Small sparkling droplets catching the golden light The debris explodes in ALL DIRECTIONS: - Some pieces fly to the RIGHT - Some pieces fly to the LEFT - Some pieces fly DIRECTLY TOWARD THE CAMERA (coming at the lens, growing larger as they approach) - Some pieces fly upward and downward The person's face remains CENTERED and CLEARLY VISIBLE throughout the explosion. Their eyes are locked on the camera the entire time, expression steady, confident, and slightly smirking or serious — a "that's how it's done" look. THEIR MOUTH STAYS CLOSED — no speaking, no shouting, no lip movement. The bat is frozen mid-follow-through. The golden backlight catches the flying droplets and glass, making them sparkle and glow beautifully. The autumn trees and warm light remain visible in the background. The explosion expands and fills MORE AND MORE OF THE FRAME as the slow motion continues. By the final moment, the debris cloud occupies a LARGE PORTION of the screen — glass, foam, liquid particles, red cap pieces are everywhere, suspended in the air, beautifully lit, with the person's confident face still centered and visible behind/within the cloud. Audio: the CRACK of the impact is deep and resonant in slow motion, followed by the slow, bassy ROAR of the explosion — shattering glass, rushing liquid, foam expanding, and carbonation hissing, all stretched out and emphasized. Cinematic, powerful, satisfying sound design. NO SPEECH, NO DIALOGUE. [00:10] END The video ENDS on this final slow-motion frame: the debris at maximum expansion, filling much of the screen, the person's face centered and staring confidently into the camera, frozen in the aftermath of the explosion. No fade, no cut to black — just hold this epic moment and end. -— GLOBAL CONSTRAINTS AND PRESERVATION Camera: FIXED POSITION throughout the entire 10 seconds. No movement, no zoom, no pan, no handheld shake. The framing stays locked. 9:16 vertical format. Continuity: This is ONE CONTINUOUS SHOT with no cuts, no scene changes, no jump cuts. The only speed change is the transition to slow motion at [00:08.5]. The person's face: stays CENTERED in the frame throughout the entire video. Always clearly visible, always in focus. The bench: A simple wooden bench (NOT a table), centered in the lower third of the frame, running horizontally. The person SITS ON THE EDGE of this bench (not standing beside it) when hitting the bottle. The bottle: glass Coca-Cola bottle (NOT plastic, NOT champagne). Classic contoured glass bottle with dark brown soda inside, red and white Coca-Cola label around the middle. SEALED with a visible red metal bottle cap (crown cap) at the top. The bottle is UNOPENED — the red cap is in place throughout until the bottle is smashed by the bat. The person holds it by the NECK and hits the BOTTOM against the bench EDGE. The bat: wooden, natural grain, classic baseball bat. Autumn atmosphere: warm golden hour light, backlight with lens flare, dense yellow/orange/red fall foliage in the background, green grass, beautiful autumn tones throughout. Lighting: professional cinematic lighting — strong backlight (rim light, lens flare), soft frontal fill light on the person's face. Clothing and appearance: taken 100% from @image1 without changes. The character is a child or teenager as shown in the reference. Audio: diegetic sound only — footsteps, movement, bottle hitting bench (THUD), bat whooshing and catching, the swing, the impact (CRACK), the explosion (slow-motion shattering and liquid spray). NO SPEECH. NO DIALOGUE. NO WORDS SPOKEN. NO LIP MOVEMENT. No background music, no subtitles. NEGATIVE CONSTRAINTS No cuts or jump cuts (except the speed ramp to slow motion at the end). No camera movement or zoom. No background music or soundtrack. No subtitles or text overlay. NO SPEECH. NO DIALOGUE. NO WORDS. The person does NOT speak, does NOT move their lips, does NOT say anything at any point in the video. Complete silence from the person. Do not change the person's appearance, clothing, gender, or age from @image1. Keep the child or teenager exactly as shown in the reference. Do not use a champagne bottle — use a glass Coca-Cola bottle with the red and white Coca-Cola label. Do not use a plastic bottle — use GLASS. The Coca-Cola bottle must be SEALED with a visible red metal bottle cap (crown cap) at the top. Do not show an already-opened bottle or a bottle without a cap. The red cap must be clearly visible. Do not use a metal bat — use wood. Do not use a TABLE — use a simple wooden BENCH, centered at the bottom of the frame. The person SITS ON THE EDGE of the bench when hitting the bottle — they do NOT stand. The person hits the BOTTOM of the bottle against the EDGE of the bench — NOT the top/neck of the bottle, and NOT against a flat table surface. The bottle does NOT open from the bench hits — it only breaks when hit with the bat. Do not show another person throwing the bat — the bat simply flies in from off-screen left.

Как сделать видео, "Богатый, но одинокий".

2026-09-14T18:29:09.700Z

1.Создаем карты персонажей (2 шт).
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

Промпт для карт персонажей:

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

RIGHT PANEL (close-up):

LEFT PANEL (back view, full body):

MIDDLE PANEL (front view, full body):

The same person, back view, in the same clothing, at full height, including the head and hair (viewed from behind). Preserve the hairstyle, body build, and clothing details.

GENERAL REQUIREMENTS:

Unified style, consistent lighting and figure scale across all panels, consistency of appearance and clothing between panels. Clean background, clear vertical dividers between panels. Realistic photographic style, high resolution.

2.Создаем видео.
Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - Качество 480p (если мажоры, то 720р или 1080р) - Соотношение сторон 16:9 - Длительность 11 сек. Загружаем КАРТЫ ПЕРСОНАЖЕЙ (парень + девушка). После вставляем промпт и вручную удаляем все "@image1", "@image2", и заново их прописываем (когда будете вводить "@" там можно будет выбрать нужный файл), так они привяжутся к промпту:

ПРОМПТ ДЛЯ ВИДЕО:

@image1 defines the MAN. From @image1 take ONLY his appearance: face, hairstyle, hair color, skin tone, facial features, and body proportions. His clothing and accessories are NOT taken from the reference — they are specified below and change between scenes.

@image2 defines the WOMAN. From @image2 take her COMPLETE appearance: face, hairstyle, hair color, skin tone, gender, body proportions, her exact clothing, and her full overall look. Do not change or redesign anything from @image2.

An 11-second vertical 9:16 cinematic short film about a wealthy but lonely man who remembers the happiness of his past. The film alternates in hard cuts between two cars: a luxurious Rolls-Royce where the man is rich, well-dressed, and heartbroken, and an old cheap car where he and his girlfriend were happy together. No dialogue.

THE CAMERA — CRITICAL, NEVER CHANGES

The camera is mounted on the car's hood, facing BACKWARD through the windshield into the cabin, looking straight inside. It is a medium shot framing the interior. The camera stays in the EXACT SAME position, angle, and framing in EVERY shot of the video — both in the Rolls-Royce and in the old car. The camera does NOT move, does NOT zoom, does NOT shake, does NOT pan. It is completely locked. The passenger seat is on the LEFT side of the frame, the driver's seat is on the RIGHT side of the frame (left-hand drive car). Only the CONTENT inside the car changes between cuts.

SCENE A — THE ROLLS-ROYCE (PRESENT DAY)

Vehicle: a Rolls-Royce luxury sedan interior. Black leather seats with orange stitching and trim, polished wood dashboard, the RR logo on the headrests, and a starlight ceiling (fiber-optic stars in the roof) clearly visible. Night outside, city lights and streetlamps passing in the background through the rear window. Cold, dark, moody colour grading — deep shadows, cool blue-black tones, muted highlights.

The man (from @image1) is driving, seated on the RIGHT side of the frame. His outfit: an impeccable black tailored suit, black dress shirt buttoned to the neck, black tie, gold luxury wristwatch on his left wrist, and a gold ring on his finger. Slicked-back neat hair. EXPENSIVE, SHARP, POWERFUL — the look of a man who has everything.

The passenger seat (LEFT of frame) is EMPTY.

His emotional state: profound loneliness, heartbreak, quiet despair. He is NOT calm — he is breaking down internally.

SCENE B — THE OLD CAR (MEMORY)

Vehicle: an old cheap 1990s budget sedan interior (like a Renault Logan or similar) — plain grey fabric seats, cracked plastic dashboard, worn steering wheel, no luxury details. Daytime, golden hour — warm low sunlight filling the cabin, soft pastel tones, nostalgic and dreamy colour grading. The whole scene glows warm and happy.

The man (from @image1) is driving, seated on the RIGHT side of the frame. His outfit: a plain black sweatshirt or hoodie — simple, casual, ordinary. Relaxed, natural, no watch or ring visible. He looks YOUNGER in spirit, joyful, laughing.

The woman (from @image2) sits in the passenger seat on the LEFT side of the frame. Her appearance and clothing are taken 100% from @image2 without any changes.

Her emotional state: pure joy, energy, love. She is the heart of this memory.

---

TIMELINE — 11 SECONDS, 9 HARD CUTS, ALTERNATING

Each cut is a HARD CUT (instant, no transition, no fade, no dissolve). The camera stays identical in every shot — only the environment, lighting, clothing, and action change. This fast alternation between "rich and miserable" and "poor and happy" is the entire point of the film.

[00:00 – 00:01] SCENE A — ROLLS-ROYCE

Night, luxury cabin, starlight ceiling. The man drives with one hand on the wheel, dressed in his black suit. He is crying silently — his face is tight with pain, jaw clenched. He brings his hand up to cover his mouth, trying to hold back sobs. Tears on his cheeks. The empty passenger seat beside him.

Audio: quiet luxury car interior hum, soft engine, faint city ambience outside, the man's shaky breathing.

[00:01 – 00:02] SCENE B — OLD CAR

Hard cut. Warm golden daylight, plain grey interior of the old car. The woman (from @image2) is in the passenger seat, singing and dancing energetically in her seat — she bounces to the music, mouth open singing, big genuine smile, hair moving. The man drives, laughing, glancing at her with pure love. Two happy people in an ordinary car.

Audio: warm engine sound, the woman's happy laughter and singing.

[00:02 – 00:03] SCENE A — ROLLS-ROYCE

Hard cut. Back to night, luxury, cold. The man is now CRYING HARD — his face twisted in grief, mouth open in a silent, agonized shout of pain, eyes squeezed shut for a moment, then open and wet. One hand still on the wheel. Complete inner devastation. The empty seat beside him says everything.

Audio: muffled interior hum, his broken breathing, a low sob.

[00:03 – 00:04] SCENE B — OLD CAR

Hard cut. Warm day. The woman turns to look at the man, still singing to him — she sings directly AT him, smiling, teasing, joyful, eyes on his face. He glances back at her and grins. Connected, in love, easy.

Audio: her singing, warm engine, laughter.

[00:04 – 00:05] SCENE A — ROLLS-ROYCE

Hard cut. Night, luxury. The man grabs fat stacks of cash (rubber-banded bundles of banknotes) and THROWS them onto the empty passenger seat — one after another, angrily, desperate. Money means nothing in that empty seat. His face is anguished and confused, as if asking himself why he has all this and nothing that matters.

Audio: the slap and flutter of banknote bundles landing on leather, his tight breathing.

[00:05 – 00:07] SCENE B — OLD CAR

Hard cut. Warm day, golden light. The longest memory beat. Both the woman and the man are laughing uncontrollably together — real, messy, joyful laughter. She throws her head back laughing, reaches over and playfully grabs his arm or shoulder. He laughs with his whole face, glancing between the road and her. They are goofing around, completely happy, completely careless, completely together. This is the emotional peak of the memory.

Audio: their overlapping laughter, the warm rumble of the old engine, the joyful chaos of two people having fun.

[00:07 – 00:08] SCENE A — ROLLS-ROYCE

Hard cut. Night, luxury. The man points with an outstretched arm at the pile of money now sitting on the empty passenger seat — his face broken, almost accusing, as if the money itself is the problem. Tears still on his face. His mouth moves silently as if he's saying something bitter he can't voice.

Audio: quiet interior, his ragged breathing, distant city sound.

[00:08 – 00:09] SCENE B — OLD CAR

Hard cut. Warm day. The woman lifts the man's hand off the steering wheel and gently, tenderly KISSES his hand — soft, loving, intimate. She looks up at him with adoration. He looks at her softly, touched, a warm smile on his face, one hand still on the wheel. A quiet, gentle, loving moment — the counterpoint to all his present-day rage.

Audio: very quiet — soft engine, faint breath, an almost silent tender moment.

[00:09 – 00:11] SCENE A — ROLLS-ROYCE (FINAL)

Hard cut. Night, luxury, cold. The man slowly lowers his hand from his face after wiping away a tear. He grips the steering wheel with both hands. His breathing slowly steadies. He stares forward at the night road through the windshield — eyes empty, hollow, absent, lost in the memory he cannot return to. The starlight ceiling glows softly above him. The empty passenger seat is visible beside him. He drives on alone.

Audio: the low hum of the luxury engine, faint city ambience, one shaky exhale, then quiet as the video ends.

[00:11] END

No fade, no cut to black — the video ends on the man's hollow stare at the night road.

---

GLOBAL CONSTRAINTS AND PRESERVATION

Camera: completely LOCKED and IDENTICAL in every single shot. Same position (on the hood, looking back through the windshield into the cabin), same medium framing, same angle throughout the entire video. The passenger seat is always on the LEFT of frame, the driver always on the RIGHT.

Editing: 9 shots separated by HARD CUTS only — no fades, no dissolves, no smooth transitions, no whip pans. Instant, sharp alternation between Scene A and Scene B.

Clothing: the man wears the BLACK SUIT look (suit, black shirt, tie, gold watch, gold ring) ONLY in the Rolls-Royce scenes, and the PLAIN BLACK SWEATSHIRT look ONLY in the old car scenes. The woman's full appearance and clothing come entirely from @image2 and appear ONLY in the old car.

Face continuity: the man from @image1 must remain clearly recognisable in every shot — same face, same hairstyle. The woman from @image2 must remain clearly recognisable.

Colour grading: Rolls-Royce scenes are cold, dark, night-time — blue-black, moody, desaturated. Old car scenes are warm, golden hour — soft, nostalgic, glowing. This contrast is essential and must be strong.

No dialogue from either character. No speech. No lip-synced words.

NEGATIVE CONSTRAINTS

No camera movement of any kind — no zoom, no pan, no shake, no push-in.

No fades, dissolves, or smooth transitions — hard cuts only.

No subtitles, no text, no captions.

Do not change the man's face or hairstyle from @image1. Do not change the woman's appearance or clothing from @image2.

Do not put the woman in the Rolls-Royce. She exists ONLY in the old car memory scenes.

Do not show the man happy in the Rolls-Royce — he is lonely and devastated there.

Do not make the old car look luxurious — it must be visibly cheap and old.

Do not fill the passenger seat of the Rolls-Royce — it stays EMPTY.

No background music unless specified. No dialogue.

🎬Как попасть в мини-сцену из американского романтического фильма 90-х?

2026-09-13T17:40:40.818Z

Нужен аккаунт в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT) Нейросети: Gpt Image 2 или Grok Imagine 2.0 для фото, Seedance 2.5 (доступен от подписки PRO и выше) для видео. Нужно: фото парня и двух девушек. Мой результат ищите в https://t.me/viviannru

1.Создаем карты персонажей (3 шт). Отправляем фото людей в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.

RIGHT PANEL (close-up):

LEFT PANEL (back view, full body):

MIDDLE PANEL (front view, full body):

The same person, back view, in the same clothing, at full height, including the head and hair (viewed from behind). Preserve the hairstyle, body build, and clothing details.

GENERAL REQUIREMENTS:

*Для девушек все тоже самое:

2.Создаем видео. Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Pro - Качество 480p (если мажоры, то 720р или 1080р) - Соотношение сторон любое (у меня 4:3) - Длительность 25 сек. Загружаем КАРТЫ ПЕРСОНАЖЕЙ. После вставляем промпт и вручную удаляем все "@image1", "@image2" "@image3", и заново их прописываем (когда будете вводить "@" там можно будет выбрать нужный файл), так они привяжутся к промпту:

Based on the uploaded reference images, strictly maintain all facial features, skin tone, hair color, hair texture, eye shape, eye color, eyebrow shape, nose bridge, lip shape, jawline, and overall physical appearance of the man in @image1, the woman in @image2, and the woman in @image3. Lock character consistency throughout the entire video. Do not alter or reinterpret their appearance — replicate exactly what is shown in the reference images, especially hair color and texture.

A chance encounter across Manhattan penthouses on a golden autumn afternoon in 1980s New York City — a woman in a white towel approaches her window after a shower, catches the eye of a handsome stranger in a tuxedo watching from the building across, leading to a phone call and a secret meeting interrupted by the arrival of his elegant companion. Cinematic 1980s American romantic melodrama aesthetic with warm golden-amber color grading, soft diffusion filters, shallow depth of field with creamy oval bokeh, gentle lighting, film grain texture, and the visual language of classic 80s luxury. 4:3 aspect ratio framing evokes the era of romantic films.

[00:00 – 00:03] Opening — Woman approaching the window

The scene opens inside a luxurious Upper East Side penthouse in 1980s New York. Daylight streams through tall arched French windows overlooking Central Park below and the Manhattan skyline in the distance — autumn trees with golden-orange leaves, classic high-rise buildings, soft afternoon light. A woman (reference @image2 for all facial features, skin tone, hair color, hair texture, eye shape, and overall appearance — replicate exactly as shown in the image) walks toward the window. She wears a white terry cloth bath towel wrapped around her. Her hair (exact color and texture as shown in @image2) falls in voluminous loose waves past her shoulders with a glamorous 80s blowout — full body and soft curls, slightly tousled. She holds a classic oval wooden paddle hairbrush with dark bristles in one hand. The camera follows her as she reaches the window and pulls aside the heavy burgundy velvet drape with gold tasseled tiebacks, revealing the view. The interior is opulent: a canopy bed with carved wooden posts and burgundy-gold bedspread, antique armchairs, a marble fireplace, crystal chandeliers, gilded console tables, and candelabras. Soft ambient sound: rustling fabric, distant traffic, birds chirping.

[00:03 – 00:20] The encounter and the phone call — romantic connection (17 seconds)

00:03–00:05: Medium shot: the woman brushes her hair and glances out the window. Her eyes widen in surprise — she gasps softly and her lips part. Cut to her point of view: across the street, in a window of the facing penthouse slightly below her eye level, stands a tall man (reference @image1 for all facial features, skin tone, hair color, hair texture, eye shape, and overall appearance — replicate exactly as shown in the image) wearing a black tuxedo jacket over a white dress shirt with the collar open in a relaxed elegant style. His hair (exact color and texture as shown in @image1) is neatly styled with natural waves swept back, classic 1980s style. He stands at his window watching her, his gaze tilted slightly upward, his expression warm and intrigued. Cut back to the woman: she quickly adjusts the towel, her expression a mix of surprise and curiosity.

00:05–00:07: The woman steps behind the burgundy velvet curtain. A beat of silence. Then she slowly peeks out from behind the drape, a small flirtatious smile on her lips. Her eyes glance back toward the man's window. The soft golden light backlights her hair.

00:07–00:09: Close-up of the man at his window. A slow, confident smile spreads across his face. He reaches beside him — a vintage cream-colored telephone with a long coiled cord sits on a small table next to the window. Without breaking eye contact, he picks up the telephone handset and holds it up toward the window so she can see it. His gesture is playful and confident: pick up your phone.

00:09–00:10: The woman watches him, her eyes widening in intrigue, then glancing toward her own telephone. A flicker of curiosity and excitement crosses her face.

00:10–00:20: The phone call — they stand at their windows, looking at each other (10 seconds). A cream or beige vintage telephone in her apartment begins to ring. The woman walks over, stands by the window, and picks up the receiver. The camera alternates between setups showing the connection between them.

Over-the-shoulder shot from behind the woman: She stands at her window, phone to her ear, looking downward across toward the man's building. Through the window, the man stands at his own window (positioned slightly lower), phone to his ear, looking upward at her.

Close-up of the man at his window, phone to ear, gazing upward with steady warm eyes:

Man (smooth, confident voice): "Can we meet? Five minutes."

Close-up of the woman at her window, phone to ear, looking downward toward him, intrigued with a faint smile:

Woman (softly, with curiosity): "Five minutes?"

Over-the-shoulder shot from behind the man: He stands at his window, phone to ear, looking upward. Through his window, the woman in her window above, phone to ear, gazes down at him.

Close-up of the man, steady and sincere:

Man: "Yes. Five minutes."

Close-up of the woman. She pauses, biting her lower lip slightly, then a knowing smile:

Woman (softly, warmly): "Maybe."

She places the receiver down. Her eyes remain on the window for a moment. Cut to the man: he places his receiver down, a satisfied smile on his face, his gaze lingering on her building.

[00:20 – 00:25] The arrival of the companion (5 seconds)

Medium shot in the man's apartment: the man turns from the window. A woman (reference @image3 for all facial features, skin tone, hair color, hair texture, eye shape, and overall appearance — replicate exactly as shown in the image) enters and places a hand on his shoulder. Her hair (exact color and texture as shown in @image3) is styled in a voluminous side-swept 80s blowout with soft waves. She wears an elegant champagne-gold silk evening gown with draped bodice. Around her neck is a statement diamond necklace with a teardrop pendant and matching earrings. Her makeup: smoky eyes, sculpted eyebrows, peach-berry lipstick with liner, contoured cheekbones.

Woman (softly): "Who was that?"

Man (smoothly): "Business."

He kisses her tenderly on the cheek. She accepts the kiss. The man straightens his jacket and walks past her toward the door. Fade to black. End at 25 seconds.

Visual style: Classic 1980s American romantic melodrama and luxury advertising aesthetic — warm golden-amber and peachy-bronze color grading with soft contrast, rich warm tones, burgundy-wine and gold accents, autumn colors through windows. Soft studio lighting with diffusers creating smooth gradations, gentle backlight from windows creating glow around hair and silhouettes, practical lights from chandeliers and candelabras. Black Pro-Mist diffusion effect for soft dreamy texture with bloom around highlights and hair. Portrait telephoto lenses 85mm–135mm for close-ups with creamy oval bokeh and shallow depth of field. Normal lenses 50–85mm for window shots showing spatial relationship between buildings. Fine 35mm film grain. Static framing on close-ups with slow push-ins; smooth handheld for following shots. Shot-reverse-shot and over-the-shoulder editing for phone conversation. 4:3 aspect ratio.

Character styling: The first woman (from @image2) wears a white terry cloth bath towel. Her hair (exact color and texture from @image2) is voluminous with 1980s blowout waves. Makeup: sculpted brows, warm smoky eyeshadow, peachy-pink blush, glossy rose-nude lipstick with liner. She holds a wooden paddle hairbrush. The man (from @image1) wears a black tuxedo jacket, white dress shirt with open collar, hair (exact color and texture from @image1) styled with waves swept back in 1980s style. The second woman (from @image3) wears a champagne-gold silk evening gown with draped bodice, diamond necklace with teardrop pendant and earrings, hair (exact color and texture from @image3) in voluminous 80s side-swept waves, bold evening makeup with smoky eyes and peach-berry lipstick.

Setting: Upper East Side Manhattan penthouses in 1980s neoclassical style. The woman's window is slightly higher than the man's window — she looks downward to see him, he looks upward to see her. Tall arched French windows overlooking Central Park below (autumn trees) and Manhattan skyline. Burgundy velvet drapes with gold tiebacks. Interior: canopy bed, antique armchairs, gilded tables, marble fireplace, crystal chandeliers, candelabras, carpeting, Persian rugs, vases with burgundy flowers. Vintage cream telephone with coiled cord beside the man's window.

Audio: Dialogue in English as scripted. Ambient soundscape: rustling fabric, distant traffic, birds, telephone ring. Background romantic instrumental music: orchestral or soft jazz (saxophone, strings, piano) in 1980s film style, swelling during phone conversation, fading as the second woman appears. No subtitles. Natural sound mixing.

Global constraints: Maintain strict consistency of all three characters' appearances as defined in @image1, @image2, and @image3 — do not alter hair color, hair texture, facial features, or skin tone from reference images. The man's window is positioned slightly lower than the woman's window — she looks downward, he looks upward. The telephone is beside the window so the man can gesture with it while remaining visible. All actions must be subtle, restrained, naturalistic. The woman does not say "Hello" when picking up the phone. Prohibit rigid cutting — all cuts motivated by character gaze or action. Prohibit objects appearing unnaturally. Prohibit additional dialogue beyond what is scripted. Prohibit subtitles. Prohibit modern digital color grading or sharp imagery — aesthetic must remain rooted in soft warm diffused 1980s romantic visual language with golden-amber palette, diffusion, backlight glow, shallow depth of field with oval bokeh, film grain, soft contrast, and slow camera movements. Evoke classic 1980s romantic film shot on 35mm with soft-focus filters and glamour lighting.

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Промпты для видео на день рождения с самолётом, машиной и шпагатом. Детская версия!

2026-08-20T15:40:27.326Z

Я все делал в Syntx.ai (https://syntx.ai/welcome/mEQ2WKuT)
Нейросети: Gpt Image 2 для фото, Seedance 2.5 (доступен от подписки PRO и выше) для видео.
Нужно: фото ребёнка и фото мамы.
Мой результат ищите в https://t.me/viviannru

1.Готовим фотографии героев:

2.При желании, можете сделать карты персонажей (можно и без них):

Take the person's appearance from the uploaded photos and create a character reference sheet, divided into three vertical panels on a neutral white or light-gray background.
RIGHT PANEL (close-up):
A close-up portrait of the person's face, cropped at the shoulders (bust shot). Face front-facing, looking directly at the camera, neutral expression, natural even lighting. Preserve all facial features and details as accurately as possible: eye shape, nose, lips, eyebrows, face shape, eye color, skin tone and texture, hair shape and color, as well as any unique features (moles, freckles, ear shape). High detail, realism, portrait quality.
LEFT PANEL (back view, full body):
The person standing at full height, front view, in their everyday clothing, arms along the body, feet shoulder-width apart. IMPORTANT: the head and face are completely absent — the figure is cropped at the neck/shoulder line, no head. Only the figure from the neck down is shown. Preserve body build, proportions, and clothing.
MIDDLE PANEL (front view, full body):
The same person, back view, in the same clothing, at full height, including the head and hair (viewed from behind). Preserve the hairstyle, body build, and clothing details.
GENERAL REQUIREMENTS:
Unified style, consistent lighting and figure scale across all panels, consistency of appearance and clothing between panels. Clean background, clear vertical dividers between panels. Realistic photographic style, high resolution.

3.Создаем видео.
Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - качество 480р (если мажоры то 720p) - Соотношение сторон 9:16 - Длительность 8 сек. Загружаем фото ребенка + мамы, после вставляем промпт и вручную удаляем все "@image1", "@image2", заново их прописываем (когда будете вводить "@" там можно будет выбрать нужный кадр), так они привяжутся к промпту.
*Для нейросети Omni flash всё тоже самое:

ПРОМПТ ДЛЯ ВИДЕО (В НАЧАЛЕ ЗАМЕНИ ЧИСЛО НА ВАШ ВОЗРАСТ):

EDITABLE SETTING — set once at the top: [AGE] = 4. Replace only this number for each new video; the on-screen text at the end must show exactly "HAPPY BIRTHDAY" on the first line and the number [AGE] alone on the second line.

Creatives Description: Based on the two uploaded images, strictly follow them in upload order: @image1 is the CHILD — take their entire appearance 100% from the reference: face, gender, age, build, skin tone, hairstyle and hair color, full outfit and sneakers exactly as in @image1 — never altered, never restyled; strictly lock this identity and outfit. @image2 is the MOTHER — take her entire appearance 100% from the reference: face, age, build, skin tone, hairstyle, full outfit and shoes exactly as in @image2 — never altered, never restyled; strictly lock her identity and outfit. Keep both faces fully independent — no feature mixing. No other characters.

One-sentence summary: An 8-second VERTICAL 9:16 cinematic birthday video on an empty runway at a breathtaking sunset: the child from @image1 smiles at the camera, runs to the mother from @image2 while turning to face the camera again, backs two small steps onto her waiting palms, and with ONE sharp upward throw she launches the child — and at the exact instant the child reaches the highest point of flight, opening into a mid-air side split, the airliner's NOSE is positioned directly above the child's head — at that precise instant time stops, the whole frame snaps to black-and-white, and glowing text "HAPPY BIRTHDAY [AGE]" appears between the child and the mother.

Specific plot description: 8 seconds, VERTICAL 9:16, smooth dynamic gimbal-like camera, never static, never shaky; location: an empty airport runway to a flat horizon; the sky is a BREATHTAKING vivid sunset — deep purple and magenta clouds blending into blazing orange, gold and crimson at the horizon; from the FIRST FRAME a large airliner with landing lights blazing approaches head-on from the far horizon, growing steadily larger the whole video — its approach is paced so that its NOSE arrives exactly above the child at the toss peak at second 6, NOT earlier.
[00:00–00:02] THE SMILE: the child from @image1 stands in the foreground CENTERED, facing the lens; the mother from @image2 waits further back near the LEFT EDGE, softly out of focus; the child looks into the camera and gives a small warm smile; the plane is still small and distant.
[00:02–00:04] THE RUN, THE TURN-BACK AND THE MOUNT — CAMERA PANS LEFT: the child turns and RUNS to the mother as the camera smoothly PANS LEFT to re-center the pair; reaching her, the child TURNS AROUND IN THE SAME MOTION — whole body rotating, not just the head — so the child now FACES THE CAMERA with their BACK to the mother; the mother has already SQUATTED DOWN and placed her open palms low in front of her; the child takes TWO SMALL STEPS BACKWARD and steps up with both feet ONTO HER PALMS, one foot per palm, standing upright facing the camera; the plane behind is now noticeably closer but still short of overhead.
[00:04–00:06] THE THROW — ONE SHARP MOTION: from the squat the mother explodes upward in ONE single sharp continuous movement — legs driving up, arms thrusting — launching the child straight up off her palms; SLOW MOTION begins as the child rises; the child opens into a clean SIDE SPLIT in mid-air, legs stretched left and right, arms spread wide, body facing the camera; the camera smoothly TILTS UP so the incoming airliner fits fully in frame, nose-on, landing lights blazing.
[00:06–00:08] THE PEAK AND THE FREEZE — CRITICAL SYNC: at second 6 the child reaches the ABSOLUTE HIGHEST POINT of the toss — and at THIS EXACT INSTANT the airliner's NOSE is positioned DIRECTLY ABOVE the child's head, the nose tip on the vertical centerline, the aircraft NOT yet passed over — nose front and center, most of the fuselage still behind it; perfect vertical symmetry: plane nose at top center, child in full split exactly beneath the nose, mother centered below with arms raised, small gap between her hands and the child; AT THIS PRECISE MOMENT TIME STOPS: the ENTIRE frame instantly converts to FULL BLACK-AND-WHITE — plane, sky, runway, mother AND child, including the child's clothes, no color anywhere — and everything FREEZES completely like a photograph, zero motion; elegant glowing white text appears in the space between the child and the mother's hands: line one "HAPPY BIRTHDAY" in thin capitals, line two the large number [AGE]; the frozen monochrome image with the glowing text holds until the end.

Global supplement: VERTICAL 9:16, 8 seconds. CAMERA: alive and smooth — LEFT PAN on the run, upward TILT on the toss, gimbal-fluid, never jerky. THE MOUNT RULE: the child's turn at the mother is a FULL BODY turn arriving back-to-mother, face-to-camera, then two small backward steps onto her palms — the mother never grabs the waist or torso, feet on palms only; the throw is ONE sharp explosive motion from squat to full extension, a short controlled height. THE SYNC RULE — most critical: the freeze fires at the EXACT instant the child is at the highest point AND the plane's NOSE is exactly overhead on the centerline — the plane must NOT have passed over yet; if anything, the nose has just arrived; never freeze with half the fuselage already past, never with the nose out of frame — the nose is fully visible at top center in the frozen frame. FULL MONOCHROME RULE: at the freeze the whole frame converts to black-and-white with no exceptions — the child too, no selective color; only the glowing white text stands out. TEXT RULE: the only text in the video is the final inscription — "HAPPY BIRTHDAY" and the number [AGE], correct spelling, appearing ONLY after the freeze. Audio: soft wind, the jet's growing rumble swelling to a majestic flyover roar at the peak, the child's happy laugh on the run, the mother's effort breath on the throw; at the freeze all sound cuts to warm quiet with a gentle shimmer as the text appears. NO dialogue, no narration, NO background music, no subtitles, no watermarks, no other text. Prohibited: the child turning only the head while the body stays — the turn is full-body within the run; the mother gripping waist/torso — palms under feet only; a slow or two-stage lift — one sharp throw; the plane passing overhead BEFORE the freeze or the freeze firing after the nose has moved past center — nose exactly overhead at the frozen instant; the nose out of frame; any color after the freeze — full monochrome including the child; text before the freeze; wrong number or misspelling; the plane appearing suddenly or from the side; the freeze drifting or breathing — a true still photo; appearance changes from references; face mixing; no other people, vehicles or birds; no music, no BGM.

Viviannru

Создаем видео, где мы танцуем на парковке.

1.Создаем карты персонажей.Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Grok Imagine 2.0 или Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

2.Создаем видео.

Для начала скачиваем видео-исходник: ССЫЛКА

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Создаем сцену из человека паука с вашими героями.

1.Создаем карты персонажей.Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

2.Создаем видео.

Для начала скачиваем 1-ую часть ролика: ССЫЛКА

Теперь скачиваем 2-ую часть ролика: ССЫЛКА

Также в Раздел Видео - Выбираем Seedance 2.5, настройки такие же, только теперь загружаем 2-ую часть исходника + фото первого человека:

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Создаем видео: "Когда купил новую машину и показываешь её друзьям..."

2.Создаем видео.

Переходим в Раздел Видео - Выбираем Seedance 2.5 - Режим Omni Reference - Качество 720p (480р, чтобы сэкономить, а если мажоры, то 1080р) - Соотношение сторон 16:9

Длительность 20 сек (если отправляете вместе с мелодией, то выбираете "AUTO").

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Ролик, Когда спрашивает мама, "Как у тебя дела?"

1.Создаем карту персонажа.Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Вы в роли морского гиганта!

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Промпты для видео: "Понимание. Прощение. Терпение. Хватит."

Как сделать видео, где вы битой разбиваете бутылку?

1.Создаем карту персонажа.Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

- Промпт с шампанским:

- Промпт с пивом:

- Промпт с колой:

Как сделать видео, "Богатый, но одинокий".

1.Создаем карты персонажей (2 шт).Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

🎬Как попасть в мини-сцену из американского романтического фильма 90-х?

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

Промпты для видео на день рождения с самолётом, машиной и шпагатом. Детская версия!

2.При желании, можете сделать карты персонажей (можно и без них):

P.S. Делюсь всеми промптами бесплатно, буду рад вашей поддержке через донаты: https://boosty.to/viviannru

1.Создаем карты персонажей.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Grok Imagine 2.0 или Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

1.Создаем карты персонажей.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

1.Создаем карту персонажа.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 - Соотношение сторон 16:9 - Качество 2к:

1.Создаем карту персонажа.
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к:

1.Создаем карты персонажей (2 шт).
Отправляем фото человека в Раздел ИЗОБРАЖЕНИЯ - Gpt Image 2 или Grok Imagine 2.0 - Соотношение сторон 16:9 - Качество 2к: