OmniHuman Demo Gallery

Main Showcase

OmniHuman supports various visual and audio styles. It can generate realistic human videos at any aspect ratio and body proportion (portrait, half-body, full-body all in one), with realism stemming from comprehensive aspects including motion, lighting, and texture details.

Talking Demonstrations

OmniHuman can support input of any aspect ratio in terms of speech. It significantly improves the handling of gestures, which is a challenge for existing methods, and produces highly realistic results.

Style Diversity

In terms of input diversity, OmniHuman supports cartoons, artificial objects, animals, and challenging poses, ensuring motion characteristics match each style's unique features.

Half-body Cases with Hand Gestures

Here are additional examples specifically showcasing gesture movements. Some input images and audio come from TED, Pexels and AIGC.

Portrait Cases

A section dedicated to portrait aspect ratio results, demonstrating OmniHuman's versatility in handling different framing styles.

Note: All videos shown on this page are originally from omnihuman-lab.github.io. To generate similar results, only a single image and audio input are required, except for demos showcasing video and combined driving signals.