High Quality Human Image Animation using Regional Supervision and Motion Blur Condition

TL;DR

This paper introduces a novel human image animation method that incorporates regional supervision and explicit motion blur modeling, significantly enhancing realism and fidelity over previous approaches.

Contribution

It proposes a new approach combining regional supervision and motion blur modeling, along with training strategies for high-resolution animation, to improve quality and realism.

Findings

Outperforms state-of-the-art methods by over 21% in reconstruction precision.

Achieves more than 57% improvement in perceptual quality (FVD).

Demonstrates superior results on the HumanDance dataset.

Abstract

Recent advances in video diffusion models have enabled realistic and controllable human image animation with temporal coherence. Although generating reasonable results, existing methods often overlook the need for regional supervision in crucial areas such as the face and hands, and neglect the explicit modeling for motion blur, leading to unrealistic low-quality synthesis. To address these limitations, we first leverage regional supervision for detailed regions to enhance face and hand faithfulness. Second, we model the motion blur explicitly to further improve the appearance quality. Third, we explore novel training strategies for high-resolution human animation to improve the overall fidelity. Experimental results demonstrate that our proposed method outperforms state-of-the-art approaches, achieving significant improvements upon the strongest baseline by more than 21.0% and 57.4% in terms of reconstruction precision (L1) and perceptual quality (FVD) on HumanDance dataset. Code and model will be made available.