AI Image to Video
Bring a still to life. Upload or select a frame, describe camera and motion, and generate a short cinematic clip.
Image to video is ideal when composition is already locked. Start from a product shot, fashion still, or storyboard frame, then add camera moves and subject motion. Generate in MiniMaxVion and keep iterating on the same board.
- Animate product, fashion, and film keyframes
- Control camera and motion with prompt language
- Stay in one studio with image and video tools
Add your still
Upload a reference image or pick a frame you already generated.
Describe motion
Specify camera path, subject movement, and pacing.
Generate the clip
Run image to video, then download or restyle the result.
Write what changes, and what stays the same
The still already defines the subject and composition, so the prompt only needs to describe time: what moves, how the camera behaves, and what must not change. Short, specific instructions keep the result close to your source.
Example: "Keep the perfume bottle fixed and sharp. The water around the base ripples outward, the splash settles in slow motion, droplets fall back into the water, and a thin streak of light slowly sweeps across the glass. Keep the framing. No camera move."
Camera moves that work well
- Locked camera with subtle motion — the safest choice for products and logos
- Slow push in — adds focus and drama to a portrait or hero shot
- Gentle orbit or parallax — shows depth in landscapes and interiors
- Tracking move — follows a walking subject in fashion and lifestyle shots
What to animate
- Product shots for ads and product pages, with light sweeps, splashes, or steam
- Fashion and portrait stills, with hair, fabric, and small subject movement
- Concept art and storyboard frames, to preview how a scene will move
- Travel and landscape photos, with drifting clouds, water, and mist
Get a better first result
Use a sharp image at the same aspect ratio as the video you want, so nothing has to be cropped or invented at the edges. Ask for one kind of motion at a time; combining a fast camera move with large subject motion is where faces and products are most likely to drift.
What is image to video?
It animates an existing still into a short AI video clip using camera and motion instructions.
What images work best?
Clear subjects with readable composition work best — product shots, portraits, and storyboard frames.
Can I edit the video afterward?
Yes. Use Video Edit to rewrite, restyle, or extend a clip you already have.
Can I set the first and last frame?
Yes. Add a first frame and, on models that support it, a last frame. The model generates the motion between the two images.
How do I stop the product or face from changing?
Say what must stay fixed, for example "keep the bottle sharp and the label unchanged, keep the framing". Limit the motion to the background, light, or a small subject movement.
Pricing · Text to Image · Image to Video · 中文