Do As I Do: Pose Guided Human Motion Copy

TL;DR

This paper introduces a novel pose-guided human motion copying method that enhances realism and temporal consistency in generated videos by combining perceptual and Gromov-Wasserstein losses, episodic memory, and sequence-based foreground generation.

Contribution

The paper proposes a new approach integrating perceptual and Gromov-Wasserstein losses, episodic memory, and sequence-to-sequence foreground generation for improved human motion copying.

Findings

Outperforms state-of-the-art methods in PSNR and FID metrics.

Achieves more realistic and temporally consistent human motion videos.

Demonstrates effectiveness across five diverse datasets.

Abstract

Human motion copy is an intriguing yet challenging task in artificial intelligence and computer vision, which strives to generate a fake video of a target person performing the motion of a source person. The problem is inherently challenging due to the subtle human-body texture details to be generated and the temporal consistency to be considered. Existing approaches typically adopt a conventional GAN with an L1 or L2 loss to produce the target fake video, which intrinsically necessitates a large number of training samples that are challenging to acquire. Meanwhile, current methods still have difficulties in attaining realistic image details and temporal consistency, which unfortunately can be easily perceived by human observers. Motivated by this, we try to tackle the issues from three aspects: (1) We constrain pose-to-appearance generation with a perceptual loss and a theoretically motivated Gromov-Wasserstein loss to bridge the gap between pose and appearance. (2) We present an episodic memory module in the pose-to-appearance generation to propel continuous learning that helps the model learn from its past poor generations. We also utilize geometrical cues of the face to optimize facial details and refine each key body part with a dedicated local GAN. (3) We advocate generating the foreground in a sequence-to-sequence manner rather than a single-frame manner, explicitly enforcing temporal inconsistency. Empirical results on five datasets, iPER, ComplexMotion, SoloDance, Fish, and Mouse datasets, demonstrate that our method is capable of generating realistic target videos while precisely copying motion from a source video. Our method significantly outperforms state-of-the-art approaches and gains 7.2% and 12.4% improvements in PSNR and FID respectively.