Pixverse Mimic

PixVerse Mimic - Motion Transfer Video Generation

What is PixVerse Mimic?

PixVerse Mimic is a motion-transfer video model that animates a still image using the motion captured in a reference video. Upload a reference clip of someone dancing, walking, or performing any action, along with a subject image of a person or animal, and PixVerse Mimic generates a new video where the subject faithfully reproduces the reference motion. Built on PixVerse's V6 architecture, Mimic extracts motion patterns frame-by-frame and reconstructs them onto a target character while maintaining visual consistency and natural movement flow.

Key Features

PixVerse Mimic delivers frame-consistent character rendering with accurate pose alignment to the source motion. It supports any human or animal subject as the target character and handles complex movements including dance choreography, athletic actions, and subtle gestures. The model accepts reference videos up to 30 seconds and outputs at 360p, 540p, or 720p resolution. Output duration matches the reference video length, giving precise control over the final result.

Best Use Cases

PixVerse Mimic excels in several creative and production workflows. Content creators can animate character illustrations or product mascots with real human motion for social media. Game developers and animators can rapidly prototype character animations from a single reference clip. Marketing teams can create engaging video content by transferring spokesperson movements onto brand characters. Dance and fitness creators can demonstrate routines using virtual avatars. The model also enables reusable motion templates — record one reference video and apply it across multiple character designs.

Prompt Tips and Output Quality

For the best results, use reference videos under 10 seconds with a single person performing clear, well-lit motion. The subject image should feature a full-body shot against a clean background for optimal motion mapping. At 540p quality, generation takes approximately 79 seconds and produces sharp, artifact-free output suitable for most use cases. Use 720p for final production renders and 360p for quick iteration. The reference video must be mp4 or mov format, under 100 MB, and the primary focus should be a person performing the motion you want to transfer.

How to Use PixVerse Mimic for Motion Transfer Videos

PixVerse Mimic transforms still images into dynamic videos by transferring motion from a reference clip. Here is how to get the best results from the API.

Getting Started

Make a POST request to https://api.segmind.com/v1/pixverse-mimic with your API key and two inputs: a reference video showing the motion you want to capture, and a subject image of the character you want to animate.

Recommended Settings

Based on testing, these parameters produce reliable results:

• •quality: 540p — the best balance of output quality and generation speed (~79s)
• •Reference video: Short clips under 10 seconds with clear, single-person motion
• •Subject image: Full-body shots with clean backgrounds work best

Parameter Guide

video_url (required): URL to your reference video. Must be mp4 or mov, under 100 MB, and 30 seconds or less. Keep videos short and focused on a single person for best motion extraction.

image_url (optional): URL to the subject image you want to animate. Works with people and animals. For optimal results, use images with clear subjects and minimal background clutter.

image (optional): Alternative to image_url — provide the subject image as a base64-encoded string for direct upload without hosting.

quality (optional, default: 540p): Choose from 360p, 540p, or 720p. Use 360p for fast previews during iteration, 540p for production-ready content, and 720p for high-fidelity final renders.

Tips for Best Results

Use reference videos with a single person performing clear, well-defined movements. Dance routines, fitness moves, and walking sequences transfer particularly well. Avoid shaky camera footage or clips with multiple people — the model tracks the primary figure.

For the subject image, full-body shots outperform close-ups. Clean backgrounds help the model isolate the subject and map motion accurately. Both photographic images and illustrated characters work well.

What file formats does PixVerse Mimic accept?

Reference videos must be mp4 or mov format, under 100 MB and 30 seconds. Subject images can be provided as a URL or base64-encoded data.

Can I animate non-human subjects?

Yes — PixVerse Mimic supports both human and animal subjects. The subject image should have a clear, identifiable figure for best results.

How long is the output video?

Output duration matches the reference video duration, from 1 to 30 seconds.

What resolution options are available?

Three quality tiers: 360p, 540p, and 720p. Higher resolutions produce sharper output but cost more per second of video.

Do I need both a reference video and subject image?

The reference video (video_url) is required. The subject image is optional — if omitted, the model generates motion-transferred video using only the reference clip.

How much does it cost?

Pricing is per second of output video: $0.05625/s at 360p, $0.0625/s at 540p, and $0.075/s at 720p.