Bidirectional Temporal Diffusion Model for Temporally Consistent Human Animation
Preprint
Abstract
We introduce a method to generate temporally coherent human animation from a single image, a video, or a random noise. This problem has been formulated as modeling of an auto-regressive generation, i.e., to regress past frames to decode future frames. However, such unidirectional generation is highly prone to motion drifting over time, generating unrealistic human animation with significant artifacts such as appearance distortion. We claim that bidirectional temporal modeling enforces temporal coherence on a generative network by largely suppressing the motion ambiguity of human appearance. To prove our claim, we design a novel human animation framework using a denoising diffusion model: a neural network learns to generate the image of a person by denoising temporal Gaussian noises whose intermediate results are cross-conditioned bidirectionally between consecutive frames. In the experiments, our method demonstrates strong performance compared to existing unidirectional approaches with realistic temporal coherenceKeywords
All Related Versions
This publication has 5 references indexed in Scilit:
- SMPL: A Skinned Multi-Person Linear Model2023
- Image Super-Resolution Via Iterative RefinementIEEE Transactions on Pattern Analysis and Machine Intelligence, 2022
- Learning temporal coherence via self-supervision for GAN-based video generationACM Transactions on Graphics, 2020
- Neural Rendering and Reenactment of Human Actor VideosACM Transactions on Graphics, 2019
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004