SFW video templates

H3 · 5 seconds · 9:16 · 768P

Photo-Booth Flash Outfit Changes

Three synchronized photo flashes transform a seated character through four distinct outfits.

The retained result includes persistent booth lettering above the character; use a text-free booth reference when rerunning it.

Make it with your character

Use your character photo with the start-frame prompt in Seedream 4.5. Then animate that new image with the H3 motion prompt. Replace Mila with your character name throughout both prompts.

In Camera, select H3 and start from your prepared image. Set 5 seconds, 9:16, and 768P. Copy the prompt below. Generation uses credits; check the estimate before starting.

Open Camera

1. Create the starting image

Use Seedream 4.5 with your character photo.

Photo-Booth Flash Outfit Changes example starting image
Show start-frame prompt
Image 1 is the sole identity reference for Mila, a fictional consenting adult age 26. Create a new photorealistic vertical 9:16 cinematic starting frame. Preserve Mila's exact recognizable face, freckles, skin tone, hazel eyes, nose, lips, light-brown hair, and adult body proportions from Image 1. Inside a vintage red-curtain photo booth, Mila sits centered on a stool in a fitted black mini dress, framed from mid-thigh up. Her hair is in a polished loose updo, hands resting clearly on her knees, direct playful gaze. The clean booth camera view fills the frame; no phone, UI, text, printed strip, or second person is visible. Exactly the described adult subject count; everyone is age 25 or older. Non-explicit starting state. No nudity, text, watermark, anatomy errors, extra limbs, duplicated people, visible phone hardware, recording UI, camera bezel, or identity drift.

2. Animate the image

Show exact example prompt
integrated_multimodal_description:
For the target video, at 0.00 seconds into the target video, <Picture 1> (from [Shot 1]) is fully referenced.
[Shot 1 | 0.00-5.00 seconds | one continuous take] Mila is a fictional consenting adult age 26. Preserve Mila's exact face, adult body proportions, hair, wardrobe, scene geometry, subject count, lighting, and camera position from <Picture 1> except for the explicitly described garment change. No cuts, no visible phone or recording UI, no new people, no identity drift, and no abrupt pose reset. The camera uses perfectly locked photo-booth framing with no visible device, no cuts, and no camera movement. At 0.00-0.80, Mila tilts her chin and prepares a pose. At 0.80-4.20, three photo flashes change Mila's outfit from black mini dress to white corset look to red satin slip to black lace bodysuit; each flash is instantaneous, the booth and seated body position remain perfectly fixed, and Mila changes only one small pose between flashes. The final black lace bodysuit has opaque coverage over intimate areas. At 4.20-5.00, she leans a few centimeters toward the lens and smiles.
overall_soundscape:
Quiet booth ambience, three synchronized shutter clicks and flash capacitor pops, subtle fabric movement, and a soft exhale.
non_diegetic_music:
An original playful minimalist beat with one accent per flash.
Example verification

Retained H3 output reviewed September 7, 2026. One approved example per template; results vary with the reference image. Model: minimax-h3-max-i2v (image-to-video).

Video SHA-256: 68a259793454ecf9686a30ea0ddd0fd3f73e3326ab61f8c9eeec09c643cc5aff

Prompt SHA-256: 2dc647d0487f0a4e81cbbbaac2c96ad22b78ad1e337d8b0ea7efb734c09c684e