Pinned Tweet
What if AI could direct a luxury beauty commercial?
From concept to storyboard to cinematic video — this entire beauty campaign was created with Vorla's AI Director. 🎬
The future of creative production is here.
That freeze-frame ad trend? Every product gets its main-character moment. ☕️🕶️📕
Flying objects, playful close-ups, cinematic slow-mo—a fun way to make products stand out.
Made with Seedance 2.5 on Vorla AI.
Free prompt below 👇
Create a 10-second photorealistic street-fashion video in vertical 9:16, with original beat-synchronized music and realistic sound effects.
CORE STORY
A stylish adult Western woman walks toward the camera while talking on her phone. A stranger running past accidentally clips her right shoulder.
The brief contact interrupts her balance and sends her coffee, sunglasses, and the book tucked under her arm outward into the foreground.
Normal motion transitions into extreme slow motion. Three tightly connected close-up shots reveal the sunglasses, coffee, and book, followed by a final composition showing the woman behind her airborne belongings.
Deliver a finished video with integrated editing, speed changes, music, and sound effects.
REFERENCE IMAGES
If reference images are supplied, preserve the woman’s identity, hairstyle, outfit, object designs, street, and lighting.
Use an airborne reference image only to guide the objects’ appearance, spacing, and height. Animate a believable release from their established starting positions.
CHARACTER AND SETTING
One fashionable adult Western woman in her mid-twenties, with fair skin, shoulder-length blonde hair, natural makeup, and realistic skin texture.
She wears an open camel blazer, a fitted ivory top, straight-leg blue jeans, and burgundy low-heeled ankle boots.
An elegant shopping street with pale stone pavement, cream boutique façades, and warm afternoon sunlight. Natural shadows and realistic reflections.
Her performance is lively and natural: a relaxed phone conversation, a brief surprised reaction to the shoulder contact, and eyes tracking her scattered belongings.
INITIAL OBJECT POSITIONS
Establish these positions clearly before the collision:
LEFT HAND:
One plain dark smartphone held against her left ear.
She keeps hold of this phone throughout the entire video. It never becomes airborne.
LEFT ARM:
One closed burgundy hardcover book, held vertically between her left upper arm and left ribs.
Its lower edge sits just above her waist.
Her upper arm presses it against her torso while her left hand holds the phone.
RIGHT HAND:
One small tan takeaway coffee cup at waist height.
One white fitted lid with a small drinking opening.
The cup is half full. No straw, sleeve, or logo.
FACE:
One pair of tortoiseshell sunglasses worn over her eyes.
Exactly four belongings in total:
One phone, one coffee cup with its lid, one pair of sunglasses, and one book.
Only the coffee, sunglasses, and book become airborne.
RUNNER
One adult stranger in a charcoal running jacket and dark trousers.
The runner approaches from behind her right side, passes close beside her, and accidentally clips the outer edge of her right shoulder with their left upper arm.
The runner’s path is toward the camera and then out through the camera-right edge.
Show a brief, light shoulder contact. No grabbing, pushing, deliberate attack, body intersection, or violent impact.
The runner continues out of frame and does not return.
TIMELINE — EXACTLY 10 SECONDS
0.0–1.8s | WALKING AND TALKING
Start immediately with a front three-quarter tracking shot, framed from mid-thigh to above her head.
The camera retreats smoothly as she walks toward it.
She holds the phone to her left ear and appears to be in a casual conversation, with a small smile and subtle natural lip movement. No intelligible spoken words are needed.
Clearly show the sunglasses on her face, coffee in her right hand, and book secured beneath her left upper arm.
At approximately 1.3s, the runner enters the background behind her right shoulder, establishing their approach before contact.
1.8–2.8s | CONTACT AND RELEASE
At approximately 2.0s, the runner’s left upper arm briefly clips her right shoulder while passing.
Her right shoulder moves slightly forward and inward. Her upper body turns a few degrees with the contact.
She takes one small balancing step. Keep her torso nearly upright, with only a subtle forward inclination of about 5–10 degrees.
Her right forearm swings slightly forward and her fingers lose their grip on the coffee cup.
Her left elbow moves slightly outward while her hand keeps the phone beside her ear, releasing the book from between her upper arm and torso.
The brief head movement lets her loosely fitted sunglasses slide off her nose.
Show the releases as consequences of the same shoulder contact:
- The coffee moves forward and slightly toward camera-right.
- The book slips outward from beneath her left arm.
- The sunglasses slide forward off her face.
The runner exits through camera-right.
Begin the transition into extreme slow motion at approximately 2.5s, as the coffee clears her fingertips.
Her expression changes naturally: eyebrows lift, lips part, and her eyes track the objects.
Her left hand retains the phone throughout.
2.8–4.4s | CLOSE-UP 1 — SUNGLASSES
Cut on the sunglasses’ forward movement to a close-up of that same pair.
Use a short, smooth camera push toward the tortoiseshell frame.
The sunglasses rotate only slightly, revealing both lenses and the naturally open temples. Preserve their rigid shape.
They remain near the woman’s upper-chest height, high above the pavement.
Focus sharply on the frame and lens reflections. Keep the same woman softly visible behind them, with no sunglasses remaining on her face.
Use shallow depth of field and a restrained stereo swish as the frame passes close to the lens.
4.4–6.0s | CLOSE-UP 2 — COFFEE
Cut to the same coffee cup, continuing its established forward-right trajectory.
Make one small tracking push toward the cup, keeping it fully inside the frame.
The cup tilts approximately 30–40 degrees.
The white lid stays firmly attached.
Show a few small coffee droplets emerging from the drinking opening. They form a short, separated arc beside the lid.
Focus briefly on the rim, drinking opening, and nearby droplets.
The cup remains around the woman’s lower-chest-to-waist height. It has barely descended because the action is in extreme slow motion.
No large splash, detached lid, deformed cup, liquid passing through the walls, or coffee covering her face.
6.0–7.6s | CLOSE-UP 3 — BOOK
Cut to the same burgundy book, just outside the woman’s left side.
Use a short diagonal camera move along the cover edge, revealing the cover texture, spine, and cream page edges.
The book remains closed and rigid, with a gentle partial tilt.
Keep it around waist height and well above her hips. It has only just slipped free.
No opening pages, fluttering paper, readable title, or sudden full rotation.
Use a soft cover-and-fabric release sound, synchronized with the visible movement.
7.6–10.0s | FINAL AIRBORNE COMPOSITION
Cut back to a medium view of the woman and all three airborne objects.
By 8.0s, establish a spacious, asymmetrical composition:
- Sunglasses in the upper-left foreground.
- Coffee cup and a few droplets in the right foreground.
- Closed burgundy book in the left middle ground near waist height.
- Woman centred behind them, nearly upright.
- Phone still held in her left hand beside her ear.
Her right hand is empty and slightly open.
Her expression shows natural surprise rather than panic.
The runner is already out of frame.
Ease the camera into a subtle forward drift. Keep small movements in her hair, blazer, and expression.
The objects continue moving slightly forward, outward, and downward. End before they approach the pavement.
No catching, ground impact, recovery walk, or return to normal speed.
SLOW-MOTION AND SPATIAL CONTINUITY
From approximately 2.5s onward, stretch a fraction of a second of physical action across the remaining screen time.
The three close-ups examine successive instants of the same accident. They do not introduce several seconds of normal-speed falling.
Keep the objects close to their original release heights, spreading forward and sideways with only slight downward displacement.
Maintain continuous trajectories across every cut.
Never reset an object to its starting position, reverse its direction, or make it jump between sides.
Preserve the same sunlight, street orientation, woman, wardrobe, and object designs.
CAMERA AND EDITING
Use a clear sequence:
Walking medium shot → visible shoulder contact → sunglasses close-up → coffee close-up → book close-up → final medium composition.
Use clean cuts matched to object movement and musical accents.
Keep the camera on the same front side of the action. Preserve screen direction and background geography.
Each close-up contains one controlled camera movement. No full orbits, whip pans, excessive handheld shake, lens distortion, or abrupt digital zooms.
The close-ups should feel connected and energetic while allowing each object to remain readable.
MUSIC
Create an original playful electro-funk soundtrack at 120 BPM in 4/4 time.
Use a tight kick, crisp dry claps, lightly swung hi-hats, warm bouncy bass, and a catchy two-note synth motif.
Confident, fashionable, lively, and slightly mischievous. No vocals or narration.
0.0–1.8s:
Start with the groove immediately. Align visible footsteps naturally with the beat.
1.8–2.8s:
At approximately 2.0s, accent the shoulder contact with one dry percussion hit and a muted clothing-contact sound.
At approximately 2.5s, use one bass accent and a short downward pitch bend to introduce slow motion.
2.8–7.6s:
Keep the underlying pulse at 120 BPM, but thin the arrangement to spacious bass, sparse snaps, and an airy version of the synth motif.
Use brief musical pickups to support the close-up changes. Keep them light rather than adding a loud hit to every cut.
7.6–10.0s:
Build gently into the final composition.
At approximately 8.0s, bring back the main bassline and two-note hook.
Resolve the phrase within the last beat, allowing a short natural decay before 10.0s.
REALISTIC SOUND EFFECTS
Keep effects synchronized to visible events:
- Soft footsteps and faint street ambience during the opening.
- Brief approaching running footsteps before contact.
- One muted fabric-on-fabric sound at the shoulder collision.
- A small cup-and-fingertip release sound.
- One smooth stereo swish following the sunglasses.
- Delicate liquid movement beside the coffee lid.
- Soft book-cover and blazer friction as the book slips free.
- Running footsteps receding as the stranger leaves.
Keep the phone conversation unintelligible and very quiet beneath the music.
No punch sound, crash, glass breaking, book impact, explosive splash, repeated whooshes, or sounds implying that anything has hit the ground.
If the generated visual timing varies slightly, align the effects with the actual contact and releases.
QUALITY AND CONTINUITY
Exactly two people during the collision: the woman and one runner.
Only the woman remains after the runner exits.
Exactly one phone, one coffee cup with one attached lid, one pair of sunglasses, and one closed book.
The phone stays in her left hand.
The coffee starts in her right hand.
The book starts beneath her left upper arm.
The sunglasses start on her face and visibly leave it once.
No duplicated belongings, extra hands, missing fingers, merged bodies, object intersections, changing clothes, floating accessories, or distorted anatomy.
No deep bend at the waist, crouching, violent fall, or reaching toward the pavement.
No added text, subtitles, logos, watermarks, or readable book title.
OVERALL FEEL
A fashionable phone call interrupted by a brief accidental shoulder bump, transformed into a lively cinematic slow-motion sequence.
Natural human reactions, clear cause and effect, three elegant moving close-ups, spacious airborne objects, catchy music, and precisely synchronized sound.
Would you scroll past this thinking it was a real outfit interview? 🐱🎤
Made with Seedance 2.5 on Vorla AI—an everyday outfit chat, with a kitten stealing the spotlight.
For fashion creators and small brands, this could be a way to test outfit stories and ad ideas before booking a shoot.🤔
The appeal is the casual, everyday feel. The test is whether the lip sync, hands and outfit details hold up.
What gave it away as AI—or did nothing stand out?
Free prompt below 👇
Prompt
Create a 25-second photorealistic street-style interview in vertical 9:16, with synchronized English dialogue and realistic street ambience.
OVERALL FEEL
A spontaneous, friendly conversation on a stylish shopping street. The interviewer and woman have just met but immediately feel comfortable chatting. Warm, playful, and lightly amused, like friends discussing an outfit. Natural smiles and small laughs, never a serious interview or a rehearsed brand advertisement.
CHARACTER AND STYLING
One stylish adult woman, approximately 27, with European features, long softly waved blonde hair, realistic skin texture, and an expressive face.
Her outfit:
• A chocolate-brown Chanel bucket hat with a small, discreet double-C emblem.
• Black oval Gucci sunglasses with subtle branding on the temples.
• A pale-blue striped ZARA cotton button-down shirt, casually tucked in, sleeves rolled to her elbows.
• Dark charcoal ZARA tailored wide-leg trousers.
• Burgundy vintage leather loafers, visibly well cared for.
• A small Louis Vuitton monogram crossbody bag resting at her right hip, with a consistent brown-and-tan pattern.
Keep the outfit fashionable, relaxed, and believable. Maintain the same garments, colours, proportions, and accessory details throughout. Brands are mentioned naturally in conversation. No floating brand labels or advertising graphics.
She securely carries one small cream-and-ginger kitten against the left side of her chest. Her left forearm supports its body and hindquarters continuously. The kitten remains comfortable, occasionally blinking, turning its head slightly, or moving an ear.
CAMERA AND INTERVIEWER
Film entirely from the interviewer’s first-person perspective using a handheld phone.
The interviewer stays off-camera. Only one hand holding a small black microphone and part of the forearm may appear near the lower edge. No interviewer face, body, reflection, or third-person angle.
The woman looks mainly toward the interviewer just beside the lens. She occasionally glances down at her outfit or affectionately at the kitten.
Use one continuous handheld take. The camera follows her gestures with small reframing movements, gentle tilts, and one modest step backward to show the trousers and shoes. Include slight human handheld movement, with short moments of stillness.
The microphone naturally follows the conversation. During downward outfit views, it may leave the frame while remaining close enough to record her voice.
SETTING AND LIGHT
A European-style shopping street with stone pavement, understated boutiques, a nearby open café doorway, distant pedestrians, and occasional slow-moving cars.
Soft overcast daylight, realistic shadows, natural colour, and believable exposure. Preserve background continuity throughout. Background pedestrians move independently without approaching or interrupting.
EXACT TIMELINE AND DIALOGUE
0.0–3.0s — THE KITTEN BREAKS THE ICE
Start in a medium shot showing her face, hat, shirt, and kitten. The interviewer approaches gently and brings the microphone toward her.
INTERVIEWER, warmly:
“Your kitten is so cute!”
WOMAN, smiling and briefly looking down at the kitten:
“Aw, thank you!”
Her response feels immediate and genuine, with a tiny amused breath.
3.0–5.5s — ASK ABOUT THE OUTFIT
INTERVIEWER, casually:
“Okay, talk me through the outfit?”
She looks back up with an easy smile and a small nod, ready to answer.
5.5–9.0s — CHANEL AND GUCCI
WOMAN:
“Chanel hat, Gucci sunnies… feeling a little fancy.”
She lightly indicates the hat brim with her free right hand, then briefly touches the outer temple of her sunglasses. Do not remove either accessory. Finish with a small self-aware grin.
The camera makes a subtle upward adjustment to include the hat, then settles near eye level. Keep her face and kitten naturally framed.
9.0–12.5s — ZARA
WOMAN, lightly playful:
“Shirt and trousers? Zara. Keeping it balanced.”
She indicates the shirt, then gestures down toward her trousers. Add a brief smile on “balanced.”
The camera follows her gesture with a gentle downward tilt, showing the shirt tucked into the waistband and the upper part of the wide-leg trousers. Keep the kitten securely supported.
12.5–16.0s — LOUIS VUITTON
WOMAN:
“The bag’s Louis Vuitton. That one’s my weakness.”
She briefly rests her right hand on the bag at her right hip, with a small amused shrug.
The camera shifts slightly toward the bag, clearly showing its shape and monogram without an extreme logo close-up. The strap remains correctly attached and crosses her torso consistently.
16.0–20.0s — THE THRIFTED SHOES
WOMAN, with a little pride:
“But the shoes? Thrifted. Best find.”
The interviewer steps back slightly and tilts down to reveal the trousers and both burgundy loafers. She slides one foot a few centimetres forward, keeping both feet grounded and her balance stable.
Show natural leather texture and the way the trouser hems fall over the shoes. No exaggerated fashion pose.
20.0–25.0s — THE PLAYFUL FINISH
The camera smoothly tilts back up to her face and kitten.
INTERVIEWER, amused:
“And your cutest accessory?”
WOMAN, glancing affectionately at the kitten:
“Oh, he’s priceless.”
They share a brief, soft laugh. The kitten makes a small head turn. End on her relaxed smile with a short natural hold.
DIALOGUE AND PERFORMANCE
Speak clear, conversational English at a comfortable pace, approximately 150–160 words per minute during spoken lines. Keep pauses brief and natural so the complete exchange fits within exactly 25 seconds.
Two distinct voices: an off-camera interviewer and the woman. Accurate lip synchronization for the woman.
Deliver the brand names clearly but casually. No exaggerated emphasis, luxury-brand sales pitch, or robotic list reading.
Use subtle facial changes, responsive eye contact, small smiles, a light shrug, and occasional quiet laughter. Avoid laughing after every sentence, constant nodding, repetitive gestures, or exaggerated expressions.
The kitten never speaks.
SOUND DESIGN
Create a realistic, layered street soundscape:
• Clear, close microphone dialogue in the foreground.
• Very faint, muffled instrumental café music drifting from the nearby doorway.
• Occasional footsteps on stone pavement.
• Indistinct pedestrian conversation at a distance.
• Soft passing traffic and one subtle vehicle pass.
• A light breeze with gentle clothing or hair movement.
• A continuous, quiet bed of city ambience.
Keep all background sounds comfortably below the dialogue. Café music should feel located inside a shop, not like an added soundtrack. No intelligible background speech competing with the interview.
Include natural breaths, tiny conversational chuckles, and subtle fabric movement. No microphone wind distortion, loud horns, sudden sound spikes, artificial whooshes, canned laughter, or dramatic musical accents.
As the camera moves toward the trousers and shoes, maintain consistent voice clarity and believable street acoustics.
CONTINUITY AND PHYSICAL REALISM
Exactly one woman and one kitten. Her left arm continuously supports the kitten. Only her right hand gestures. She never switches the kitten between arms.
Maintain the same face, hair, hat, sunglasses, clothing, bag, shoes, kitten markings, and street layout throughout.
Keep the kitten’s weight and contact believable. No floating paws, fused fur and fingers, duplicated limbs, or changing kitten size.
The crossbody strap must not pass through the kitten, her arm, or her clothing. Accessories remain physically stable when touched.
When the camera returns to her face, preserve her identity and the exact placement of her hat and sunglasses.
OUTPUT
Exactly 25 seconds. A complete interview with integrated dialogue, camera movement, ambient sound, and a natural ending, ready to share without requiring additional transitions or sound editing.
No subtitles, added text, brand title cards, watermarks, dramatic filters, slow motion, or decorative transitions.
The close-ups are where the diamonds show off. 💎✨
Deep blue sapphires, flashes of diamond fire, and slow camera moves—a luxury jewelry ad made with Seedance 2.5 on Vorla AI.
Free prompt below 👇
Prompt
Create a 10-second photorealistic luxury jewelry advertisement in horizontal 16:9, with four carefully composed shots, refined model movement, realistic gemstone reflections, and integrated background music.
VISUAL DESIGN
An adult female fashion model, approximately 28 years old, with a sleek dark-brown low bun, natural skin texture, understated makeup, and a calm, confident expression.
She wears a champagne-colored satin off-the-shoulder evening gown. Keep the neckline low enough for the entire necklace to remain clearly visible.
Jewelry: a coordinated platinum and deep-blue sapphire set.
— Necklace: a fine platinum chain with one vertically oriented oval sapphire pendant, surrounded by a single delicate halo of small round white diamonds.
— Earrings: matching smaller oval sapphire drops, each surrounded by a single diamond halo and suspended from one small round diamond stud.
The sapphires are transparent, richly blue, and realistically faceted. Preserve exactly the same jewelry design, proportions, stone color, and setting throughout every shot.
If a reference image is supplied, use it as the visual source for the model, gown, and jewelry. Preserve those details across all camera angles.
SETTING AND LIGHTING
A seamless midnight-blue studio background with a subtle tonal gradient. A large soft key light creates flattering skin tones and smooth satin highlights. Restrained side and rim lighting separates the jewelry and shoulders from the background.
Use small, controlled highlights that travel naturally across the gemstones and metal as the model or camera moves. Diamond sparkle should be occasional and physically motivated, never a constant glitter overlay.
TIMELINE — EXACTLY 10 SECONDS
0.0–3.0s | MODEL AND COMPLETE JEWELRY SET
Medium close-up from the upper chest to above the head. Both earrings and the complete pendant are visible.
The model slowly turns her face a few degrees toward the camera, gently lifts her chin, and settles into direct eye contact. Include subtle breathing and one natural blink.
The camera makes a very slow, restrained push forward. The necklace rests naturally against her skin.
3.0–5.5s | SAPPHIRE PENDANT MACRO
A clean cut to a macro close-up of the same pendant resting against the model’s upper chest.
Keep the full oval sapphire, its single diamond halo, and the chain attachment visible.
The camera makes a tiny lateral move, revealing believable depth in the gemstone and a soft traveling reflection along the platinum setting.
Maintain sharp jewelry detail, with the surrounding skin softly out of focus. No hands enter the shot.
5.5–7.5s | MATCHING EARRING DETAIL
A clean cut to a close-up of one ear, the matching earring, and a small portion of the cheek and neck.
The hair remains neatly behind the ear.
A very small head movement creates one gentle, damped sway of the earring. Its stud remains attached to the earlobe, and the drop moves naturally from its connecting joint.
Keep the oval shape, diamond halo, and blue color consistent with the opening shot. Use a steady camera.
7.5–10.0s | FINAL MODEL PORTRAIT
A clean cut back to a medium close-up showing the model’s face, both earrings, the complete necklace, and the champagne satin neckline.
She makes a slight shoulder adjustment, returns her gaze to the lens, and forms a subtle, relaxed smile.
Hold the final composition steadily for the last second so the jewelry can be clearly appreciated. Keep the pendant unobstructed.
EDITING AND PACING
Deliver a finished advertisement with the four shots in the specified order.
Use precise, clean cuts between the portrait and detail shots. Match the model, jewelry, lighting, and background across cuts.
Keep movement slow and intentional within each shot. Avoid dissolves, morphing transitions, whip pans, speed ramps, or sudden zooms.
Do not show the jewelry being removed or changing into a separate tabletop arrangement.
INTEGRATED MUSIC AND SOUND
Generate a continuous original instrumental soundtrack: approximately 78 BPM, a sparse felt-piano melody, warm sustained strings, and a very soft ambient bed.
The mood is intimate, elegant, and quietly confident.
Place subtle musical accents at the cuts around 3.0, 5.5, and 7.5 seconds. Use one faint, delicate bell-like accent for the pendant reveal, blended into the music.
Keep the soundtrack smooth across every cut and resolve gently by 10 seconds.
No vocals, spoken words, heavy drums, dramatic bass hits, repeated sparkling chimes, or transition whooshes.
QUALITY AND CONTINUITY
Realistic skin pores, natural eyes and blinking, believable satin folds, accurate jewelry attachment, and stable facial identity.
Exactly one necklace and one matching pair of earrings.
Maintain the same oval sapphire shapes, single diamond halos, chain thickness, and platinum settings throughout.
Do not enlarge or redesign the jewelry in macro shots.
No floating pendant, fused chain links, duplicated gemstones, detached earrings, exaggerated swinging, flickering facets, plastic skin, or excessive beauty smoothing.
No added text, captions, logos, or watermarks.
OVERALL FEEL
An intimate, polished luxury jewelry campaign: a living model, clear product detail, restrained movement, and believable sparkle.
Prompt
A realistic live-action subway carriage with cinematic photography quality. The seating arrangement is fixed from left to right: a golden retriever puppy, a small border collie puppy, and a blonde foreign little girl. Both dogs are young puppies.
Character outfit details:
1. Left — Golden retriever puppy: Wearing a red baseball cap, a black suit vest, and a red-and-white striped tie. A beige canvas tote bag is placed on the floor beside the puppy.
2. Middle — Small border collie puppy: Wearing a dark green oversized bow on its head, a white ruffled blouse, a light green pleated JK-style skirt, a soft plaid tie, and a small green crossbody bag.
3. Right — Blonde little girl: A cute and gentle-looking blonde child with a soft innocent face, sitting quietly beside the border collie puppy.
15-second precise continuous action sequence
One single uninterrupted shot, no cuts, no skipped actions:
0-3s: Wide horizontal shot. The camera slowly pushes forward. The golden retriever puppy tilts its head upward and opens its mouth wide, lazily yawning. The subway carriage gently sways with the movement of the train.
3-6s: The little girl slowly tilts her head and fully rests her head on the shoulder of the border collie puppy, which was originally resting with its eyes closed.
6-9s: The border collie puppy suddenly opens its eyes wide in surprise, turns its head toward the golden retriever puppy on the left, makes a soft, baby-like puppy whimper, and gently raises its front paw to tap the golden retriever’s arm, signaling something.
9-12s: The golden retriever puppy understands the signal. It stands up, carefully grabs a soft thin blanket from inside the canvas tote bag with its mouth, completely leaves its original seat, and walks toward the little girl.
12-15s: The golden retriever puppy stands beside the little girl, gently stretches out the blanket with its mouth, and carefully covers the sleeping girl with the entire blanket. The golden retriever and border collie puppies lower their heads and look at the girl warmly. The scene freezes into a peaceful, heartwarming final frame.
Environment & visual style: Cool-toned subway ambient lighting mixed with soft warm facial light. Nighttime city lights create tiny reflections through the train windows. Shallow depth of field with a softly blurred background.
Visual style: Kodak Portra 400 vintage film texture, warm low-saturation color grading, healing vlog atmosphere, ultra-realistic 8K quality, detailed puppy fur texture, realistic fabric details, natural child skin texture, soft cinematic shadows, gentle slow-motion feeling.
No text, no watermark, no mobile interface UI. Clean and minimal subway carriage, no unnecessary objects.
Negative Prompt: Blurry image, out-of-focus subjects, distorted animal faces, deformed limbs, extra legs or paws, unnatural anatomy, messy unrealistic fur, harsh flash lighting, excessive noise, ugly or awkward composition. No subtitles, no watermark, no short-video buttons, no phone screen interface. No cartoon style, no 3D plastic appearance, no strong camera shake. No dirty or abandoned subway carriage, no strange color distortion. No East Asian-looking girl, no adult dogs, no missing puppy outfits or accessories, no deformed clothing. No broken action sequence, no incomplete movements. The golden retriever must leave the seat and walk to the girl. The blanket must be fully placed over the sleeping girl. No incorrect seating order. Dogs must wear all specified clothing and accessories. The final scene must show the girl completely covered by the blanket.
Prompt
Create a complete 15-second photorealistic fashion commercial in horizontal 16:9.
CONCEPT
A first-person virtual wardrobe controlled by two visible bare hands. The RIGHT hand swipes through floating outfits and selects the final look. Reveal a model wearing that exact outfit, then show several concise fashion-detail shots before a full-body finish.
Prioritise clear gesture-to-action timing, fluid transitions, clothing consistency, lively sound design, and realistic movement.
STUDIO
A seamless misty blue-grey cyclorama studio with a matte floor, soft directional lighting, and gentle contact shadows.
Keep the background, lighting, and colour grade consistent across all shots. Realistic fabric texture and natural skin detail. No furniture, clothing racks, interface graphics, or on-screen text.
FIRST-PERSON HANDS
Two bare adult hands appear naturally in the lower foreground:
LEFT hand on the lower-left side.
RIGHT hand on the lower-right side.
Show the backs of the hands with fingers pointing towards the floating outfits, as if the viewer is interacting with an invisible wardrobe interface. Both forearms extend naturally out of the bottom of the frame.
Natural skin, short clean nails, no gloves or jewellery.
The LEFT hand stays relaxed and makes only subtle supporting movements. It never triggers an outfit change.
The RIGHT hand performs exactly two right-to-left swipes, followed by one forward selection tap.
Keep the hands anatomically correct, clearly separated, and below the main clothing display. They never grab the garments or cross each other.
FLOATING WARDROBE
Each outfit floats as a coordinated invisible-mannequin ensemble with believable wearable volume. No visible mannequin, body, head, hangers, or supporting wires.
Keep the outfits at the same central position, scale, and height. Garments and accessories remain clearly separated.
LOOK 1
A cherry-red short-sleeved A-line midi dress, black low-heeled pumps, and a small structured black handbag.
LOOK 2
An ivory sleeveless knit top, charcoal-grey wide-leg trousers, and white low-profile sneakers. No handbag.
LOOK 3 — FINAL SELECTED OUTFIT
A forest-green long-sleeved blouse with a pointed collar and matte woven fabric.
High-waisted ivory tailored trousers with clean front pleats.
A narrow dark-brown leather belt with a small rectangular gold buckle.
Dark-brown leather loafers.
No jacket, jewellery, handbag, logos, or prints.
Preserve every detail of Look 3 across the floating display, model reveal, close-ups, and final shot. Never change its colour, cut, collar, belt, buckle, trouser shape, or shoes.
MODEL
One adult woman with a chin-length dark bob and natural makeup. Realistic proportions, relaxed confidence, and subtle facial expression.
She appears only after the selection, already wearing the complete final outfit. No gradual body formation or visible dressing process.
TIMELINE — EXACTLY 15 SECONDS
0.0–2.0s | FIRST-PERSON OPENING
Start immediately with Look 1 floating in the centre and both hands visible in the foreground.
The right hand performs one smooth right-to-left swipe. Look 1 slides left as a complete ensemble, directly responding to the gesture.
The left hand remains relaxed at the lower-left edge.
Use a short, directional fabric-like whoosh synchronised with the slide.
2.0–4.0s | SECOND LOOK
Look 2 slides in from the right and settles briefly at centre. Keep a clear gap between the outgoing and incoming outfits.
The right hand resets with relaxed fingers, then makes a second deliberate right-to-left swipe.
Look 2 slides out left.
The reset movement must not trigger a change.
Add a second short whoosh on the musical beat.
4.0–6.0s | SELECT THE FINAL LOOK
Look 3 enters from the right and stops at centre.
The right index finger points towards the green blouse and makes one small forward tap.
The entire outfit responds with a subtle unified forward movement, then settles.
Add a crisp, restrained selection tick with a soft low musical accent. No glowing button or visible interface.
The right hand then opens and approaches the lens, briefly filling the entire frame. The left hand drops out of view.
6.0–7.5s | MODEL REVEAL
Hide one clean match cut behind the right palm covering the lens.
As the palm clears the camera, reveal the model wearing the exact selected outfit, standing where the floating ensemble had been.
Match the clothing silhouette, screen position, colours, and studio lighting across the cut.
Both foreground hands now exit completely and never reappear.
Show a clear full-body view. The model gives a small confident smile and begins a gentle quarter-turn.
7.5–9.0s | COLLAR AND FABRIC DETAIL
Cut on the model’s turning movement to a close-up from the shoulders to the upper chest.
Show the pointed collar, clean seams, and matte texture of the forest-green blouse.
Her shoulder movement creates a small, believable fabric fold.
Use a subtle lateral camera move rather than a rapid orbit.
Add a delicate fabric rustle beneath the music.
9.0–10.5s | WAIST AND BELT DETAIL
Use a clean match-on-action cut following the same turning direction to a close-up of the waist.
Show the green blouse tucked neatly into the ivory trousers, the narrow brown belt, the rectangular gold buckle, and the trouser pleats.
The model makes a small weight shift, allowing the fabric to settle naturally.
Keep her hands outside this close-up to avoid unnecessary finger movement.
Add a quiet cloth movement and a short percussive musical accent. No exaggerated metallic buckle sound.
10.5–12.0s | TROUSER HEM AND SHOE DETAIL
Cut on the next beat to a low close-up showing the ivory trouser hems and brown loafers.
The model takes one small, controlled step onto the matte studio floor.
Show realistic trouser movement, shoe shape, and contact shadows.
Synchronise a soft leather-shoe footstep exactly with floor contact.
No slow motion or dramatic camera spin.
12.0–15.0s | FULL-BODY FINISH
Cut as the foot plants to a matching full-body three-quarter view, completing the same step.
The model settles into a relaxed fashion pose with one foot slightly forward and arms resting naturally.
She looks towards the camera with a subtle smile.
Use a slow, minimal camera pullback to show the entire outfit, including the shoes, then let the camera come to rest.
Hold the finished pose for the final second with natural breathing and slight fabric settling.
Finish on a clean musical resolution.
EDITING AND TRANSITIONS
Use different transitions only where motivated by the action:
• Smooth horizontal carousel movement during outfit selection.
• One palm-covered match cut for the model reveal.
• Clean match-on-action cuts between the model and garment details.
• A footstep match cut back to the final full-body view.
Maintain the model’s movement direction and body orientation across cuts.
No decorative spinning transitions, glitch effects, particle dissolves, random zooms, excessive motion blur, or strobing flashes.
MUSIC AND SOUND
Compose original upbeat fashion-house music at approximately 120 BPM in 4/4 time.
Use a rounded, bouncy bassline, a clean kick drum, crisp claps, light hi-hats, and a memorable short plucked-synth motif.
Keep the opening rhythm clear for the wardrobe gestures. Add a small musical lift at the model reveal, followed by distinct accents for the three detail shots. Resolve the phrase naturally at the end.
Sound effects must support the visible actions:
Two short directional whooshes for the two swipes.
One precise selection tick for the forward tap.
One soft close whoosh as the palm covers the lens.
Subtle fabric movement during the blouse and waist shots.
One gentle shoe-on-floor contact during the step.
Keep sound levels balanced and polished. No constant whooshing, loud impacts, dialogue, narration, or lyrics.
QUALITY AND CONTINUITY
Exactly two foreground hands during selection: one left and one right.
Only the right hand controls the wardrobe.
Exactly three outfits, shown in order, and one model.
No foreground hands after the model reveal.
Maintain the selected outfit exactly across every shot.
No duplicated garments, extra fingers, floating limbs, visible mannequins, intersecting accessories, fabric melting, face changes, or background shifts.
Generate the complete video with integrated editing, music, and synchronised sound effects, requiring as little additional post-production as possible.
No subtitles, logos, text overlays, end cards, or watermarks.
Create a 15-second photorealistic fashion e-commerce video in vertical 9:16 format. Generate the visuals, original music, and synchronised sound effects together as one finished video.
CONCEPT
One expressive adult woman starts standing beneath a fixed overhead camera, smoothly lies down on an ivory rug, and claps twice. The second clap cues a fan blade to begin rotating.
The blade passes close to the camera twice, completely concealing the image during each outfit change. Reveal three distinct outfits.
Between transitions, she models the clothes with natural expressions, elegant gestures, and a gentle side roll.
The atmosphere is warm, creamy, stylish, playful, and energetic.
CAMERA — FIXED OVERHEAD THROUGHOUT
Use one stationary camera mounted at ceiling level, pointing vertically downward at 90 degrees to the floor.
Keep exactly the same camera position, viewing direction, focal length, and framing for the entire video.
Compose the rug vertically, with enough space to show her complete reclining body, including her shoes, and allow small arm movements and a gentle side roll.
At the beginning, she is standing beneath the camera. Show the natural top-down view of her head, shoulders, and foreshortened body. A frontal standing view is not required.
She briefly looks upward at the camera, then bends her knees, sits down with hand support, and smoothly reclines onto her back.
Capture this entire action continuously from the same overhead camera. As she lowers herself toward the floor, allow her apparent size and silhouette to change naturally with perspective. Do not compensate with a zoom or artificial body resizing.
Once she lies down, show her face and the front of her complete outfit. Her head points toward the top of the frame, and her feet toward the bottom.
No camera cuts during the opening movement. No frontal or side camera, orbit, pan, tilt, zoom, or reframing.
Only the two outfit changes may use hidden cuts, fully concealed behind the fan blade.
CHARACTER AND PERFORMANCE
One adult woman, approximately 25–30 years old, with shoulder-length dark brown hair, natural makeup, and realistic skin texture.
Maintain the same face, hairstyle, body proportions, and identity throughout.
She is charismatic and comfortable in front of the camera, with natural blinking, breathing, small head movements, and changing expressions:
- An inviting smile while standing.
- A playful, anticipatory expression before clapping.
- A confident half-smile in the tailored outfit.
- A bright, relaxed smile in the final satin look.
While reclining, she actively presents the clothing through small, purposeful movements:
- Trace the outside seam of the trousers.
- Lightly hold a blazer lapel.
- Roll partly onto one side to show the waist silhouette.
- Smooth the satin skirt and release a small fold.
Keep movements fluid and realistic. Her hair rests on the rug, and clothing follows gravity.
She may change poses between transitions. Only during each hidden outfit cut must her body position match precisely.
No frozen mannequin pose, exaggerated contortions, mechanical motion, or sexualised framing.
SETTING AND DECOR
A minimalist cream-toned room with pale oak flooring and a large plush ivory rug.
Place a few decorative details around the perimeter:
- A rounded boucle cushion near one upper corner.
- A folded oatmeal throw at the rug’s edge.
- A small travertine side table with pale apricot flowers, partially visible at the outer edge of the composition.
Keep the centre clear for the model.
Decorations must not cover the outfits or interfere with her movement.
Use soft daylight from frame left, warm neutral colours, subtle peach and butter-yellow accents, gentle shadows, and fresh, luminous exposure.
Keep the background, lighting, and decorations consistent throughout.
THREE OUTFITS
LOOK 1 — RELAXED DAYWEAR
An ivory fitted short-sleeve knit top tucked into camel high-waisted wide-leg trousers, with cream leather loafers.
LOOK 2 — TAILORED CHIC
A pale butter-yellow blazer over a matching knee-length tailored dress, with beige slingback heels.
LOOK 3 — EVENING SATIN
A champagne satin midi dress with an asymmetric neckline and delicate gold-toned heels.
Show distinct fabric textures: knit, structured tailoring, and satin.
Keep garments opaque, correctly fitted, and physically believable in standing and reclining positions.
Only the clothing and shoes change between looks.
No handbags, handheld props, logos, or added text.
FAN POSITION AND MOVEMENT
The spatial arrangement from top to bottom is:
1. Fixed overhead camera.
2. Fan blade in the immediate foreground.
3. Woman on the rug well below.
Maintain a clear physical separation between the fan and the model. The blade never touches her body, hair, clothing, or decorations.
Use one visible broad matte-ivory blade with a narrow caramel stripe near its tip. It remains attached to a rotor hub outside the crop.
The blade rotates in a plane parallel to the floor. Position the camera off-centre from the rotor so the marked blade crosses the lens once per revolution.
Because it passes very close to the camera, the blade appears as a broad, softly blurred foreground shape that briefly covers the entire image.
Its edge follows a curved rotational path from upper left toward lower right. Repeat the same direction and geometry for both transitions.
Do not show a floating blade, straight sliding panel, curtain, or digital transition.
The blade starts stationary outside the frame.
It begins rotating after the second clap.
It crosses the frame twice, revealing new outfits at 6.0 and 10.0 seconds.
After the second pass, smoothly slow and stop the rotor with the blade outside the frame before a third pass occurs.
CLAPS AND OUTFIT TRANSITIONS
Show two distinct hand claps at 3.0 and 3.5 seconds.
Her palms visibly meet and separate for each clap. Synchronise the clap sounds exactly with contact.
The second clap cues a faint motor-start sound and the fan begins rotating. No magical particles or lighting effects.
For each outfit change:
- She briefly settles into a simple reclining pose before the blade arrives.
- The blade fully conceals the entire image.
- Change the outfit through an invisible hard cut during complete occlusion.
- The blade clears to reveal the next outfit already complete.
- She resumes natural modelling gestures.
Match her face, head, shoulders, hands, hips, and feet across each hidden cut.
No visible dressing, melting fabric, gradual garment transformation, flashing exposed skin, or duplicated bodies.
MUSIC AND SOUND — ENERGETIC AND SYNCHRONISED
Create an original upbeat instrumental disco-house fashion track at 120 BPM in 4/4 time.
Use a punchy but comfortable kick, warm bouncy bass, bright rhythmic guitar or playful synth chords, crisp hi-hats, and light percussion.
The mood is joyful, confident, fashionable, and lively.
No vocals, narration, sleepy lounge music, aggressive EDM drops, or orchestral impacts.
Start the music immediately at 0.0 seconds.
Maintain a steady tempo: one beat every 0.5 seconds and one bar every 2 seconds.
Leave a short rhythmic space around the real hand claps at 3.0 and 3.5 seconds so they are clearly audible.
Bring in the fuller bass-and-drum groove at 4.0 seconds.
Synchronise the two outfit reveals with strong musical downbeats:
- LOOK 2 becomes visible at 6.0 seconds.
- LOOK 3 becomes visible at 10.0 seconds.
The accent must coincide with the REVEAL of the outfit, not the start of the blade sweep.
Add a short, soft air-swish following each blade pass. Let it resolve into the reveal beat. Keep motor noise unobtrusive.
Maintain the energy through the final outfit. Use a short closing musical phrase during 14.0–15.0 seconds, resolving naturally without an abrupt cutoff.
EXACT 15-SECOND TIMELINE
0.0–0.5s | STANDING HOOK
From the fixed overhead camera, show the woman standing on the rug in LOOK 1.
She briefly tilts her face upward and gives an inviting smile.
Music starts immediately.
The fan blade remains stationary outside the frame.
0.5–3.0s | SMOOTH RECLINE
In one continuous movement, she bends her knees, lowers into a supported seated position, and reclines onto her back.
No camera cut, fall, jump, teleportation, or unnatural acceleration.
Her hair settles onto the rug.
As she finishes reclining, she raises her hands above her chest, ready to clap.
3.0–4.0s | CLAP TO START
First clap at 3.0 seconds.
Second clap at 3.5 seconds.
She looks toward the camera with a playful expression.
The second clap cues the fan to start.
She separates and lowers her hands naturally.
The fuller music groove enters at 4.0 seconds.
4.0–5.5s | LOOK 1
She gently traces one hand along the outside seam of the trousers, showing their wide-leg silhouette.
Add a small head turn and a natural smile.
Keep the outfit and shoes visible.
By 5.5 seconds, she settles onto her back with her arms beside her body, ready for the transition.
5.5–6.0s | FIRST FAN WIPE
The marked blade sweeps across the lens and briefly conceals the entire image.
Switch to LOOK 2 while fully hidden.
Reveal LOOK 2 on the musical downbeat at exactly 6.0 seconds.
6.0–9.5s | LOOK 2
She lightly holds a blazer lapel and gently rolls her shoulders and hips about 20–30 degrees onto one side.
Show the waist and blazer structure while keeping her face turned toward the camera.
Give her a confident half-smile.
Do not turn fully onto her stomach or hide the front of the dress.
She smoothly returns to the same simple reclining pose by 9.5 seconds.
9.5–10.0s | SECOND FAN WIPE
The same marked blade repeats the curved foreground sweep.
Briefly conceal the entire image and change to LOOK 3.
Reveal LOOK 3 on the musical downbeat at exactly 10.0 seconds.
10.0–13.5s | LOOK 3
She gently smooths the outer skirt with one hand, drawing the satin outward a few centimetres and releasing it.
Show a small fold catching the light and settling naturally.
Her other arm extends in a relaxed editorial pose without covering the neckline.
She turns her head slightly, then looks back toward the camera with a bright smile.
Keep the dress and shoes visible.
The fan slows and stops outside the frame.
13.5–15.0s | FINAL HOLD
She settles into a relaxed final pose with softly asymmetric arms.
Maintain natural breathing, blinking, and warm eye contact.
Keep the final outfit clearly visible.
Let the music resolve naturally.
End on the model, with no black frame, end card, or additional fan pass.
FINAL QUALITY AND DELIVERY
Exactly one woman, three outfits, two claps, and two full-frame fan-blade wipes.
Deliver one complete 15-second vertical 9:16 video with original music and sound effects already synchronised, ready for direct use without external trimming, retiming, or music alignment.
Prioritise natural human performance, clear clothing presentation, a fixed overhead camera, continuous standing-to-reclining movement, stable identity, and clean beat-matched outfit reveals.
If necessary, simplify gestures rather than accelerating them unnaturally.
No distorted anatomy, extra fingers, changing faces, floating hair, duplicated garments, unstable backgrounds, exposure flashes, camera movement, subtitles, watermarks, or decorative effects.
No costume? Just take your head off. 🎈🎃
Turn your photo into a Halloween balloon-head portrait with Vorla AI.
Log in for free credits. Upload a photo. Copy the free prompt. Generate.
Link + free prompt below 👇
Show us what happens when you let it go to your head.
🤖 Made with AI
Free prompt
Edit @Image 1 with minimal, local changes: the person’s head is magically absent from their shoulders, and they cradle their own head snugly against their chest in both arms, like hugging a cherished doll. Add recognisable Halloween makeup to the held face.
PRESERVE THE ORIGINAL PHOTO
Keep the original framing, camera angle, perspective, background, lighting, exposure, white balance, colour palette, contrast, saturation, and photographic texture unchanged.
Preserve the person’s original stance, torso angle, shoulder position, clothing, accessories, and body proportions. Adjust only the arms and hands as needed to create the cradling embrace. Do not redesign the pose or change the overall composition.
No new background, cinematic colour grading, coloured lighting, fog, or Halloween decorations.
CRADLING EMBRACE — THE MAIN ACTION
Exactly one person and one head. Remove the original head from above the shoulders and place that same head snugly against the upper torso, just below the collarbones.
The head rests within the curve of both arms, close to the body. One forearm supports it from underneath, with the palm cupping the lower jaw. The other arm wraps gently around it, with the hand resting against the side or back of the head.
Keep the elbows close to the torso. The pose should clearly read as affectionately hugging and cradling the head against the chest.
Do not hold the head out towards the camera, display it on two open palms, or carry it at waist level. No visible gap between the held head and the chest.
Keep the held face visible, with neither hand covering the eyes, nose, or mouth. Preserve the head’s natural size, recognisable features, hairstyle, and original expression. Turn it only as much as necessary to keep the face readable.
Use believable arm anatomy, gentle fabric compression, and subtle contact shadows. Conceal the underside of the held head naturally with the supporting arm, hand, and hair. No exposed anatomy.
SMOOTH, SEALED NECK
Keep a short neck in its original position above the clothing neckline. Its top is seamlessly closed by smooth, continuous skin with a gently rounded surface, matching the surrounding skin tone, texture, and lighting.
No indentation, hollow, opening, cavity, dark recess, wound, blood, scar, stitches, cut edges, or exposed interior. The effect is a clean supernatural illusion, never an injury.
HALLOWEEN MAKEUP — HAUNTED PORCELAIN DOLL
Transform the held face with clearly recognisable Halloween haunted-doll makeup while preserving the person’s identity:
A thin veil of pale porcelain foundation that retains realistic skin texture and facial contours.
Smoky charcoal and muted plum eyeshadow around the eyes.
Fine black painted lower lashes for an eerie doll-like effect.
Small, softly defined burgundy blush circles on the cheeks.
Deep wine-red lips with slightly extended, delicately painted mouth corners.
A small black spiderweb painted at one temple.
All details must look like makeup applied to intact skin. No actual cracks, cuts, wounds, stitched flesh, hollow eye sockets, or glowing eyes. Do not replace the person’s face with a mask or change their facial structure.
Keep the original expression; let the Halloween makeup and impossible head placement create the eerie effect.
LOCAL EDITING AND REALISM
Reconstruct the background where the original head was removed using the surrounding scene. Match the lighting on the held face to its new position against the chest.
Preserve everything outside the necessary edit areas as closely as possible. Do not crop, expand, zoom, change the body shape, or alter the original photo’s colour tone.
Photorealistic, stylish, uncanny, and non-graphic.
No duplicate head, extra face, extra arms, extra fingers, floating hands, intersecting limbs, balloons, strings, blood, wounds, scars, text, logos, or watermarks.