DreamActor-M1 Demo

Motion Guidance

This demo showcases our hybrid motion guidance system combining facial expressions, 3D head modeling, and body skeletons. Notice how the animation maintains precise control while preserving the subject's identity throughout the sequence.

Scale Adaptability

Here we demonstrate our model's ability to handle different scales through progressive training. The animation smoothly adapts from portrait to full-body views while maintaining consistent quality and detail.

Temporal Coherence

This example highlights our enhanced temporal coherence achieved through integrated sequential motion patterns and visual references. Note the smooth transitions and consistent appearance across frames.

Portrait Animation

This demonstration shows our model's capability in portrait animation, highlighting the fine-grained control over facial expressions and head movements while maintaining high fidelity and temporal consistency.

Note: All videos shown on this page are originally from grisoon.github.io/DreamActor-M1

To generate similar results, only a single image and audio input are required, except for demos showcasing video and combined driving signals.