Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2403.16510 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909149328048128 |
|---|---|
| author | Huang, Ziyao Tang, Fan Zhang, Yong Cun, Xiaodong Cao, Juan Li, Jintao Lee, Tong-Yee |
| author_facet | Huang, Ziyao Tang, Fan Zhang, Yong Cun, Xiaodong Cao, Juan Li, Jintao Lee, Tong-Yee |
| contents | Despite the remarkable process of talking-head-based avatar-creating solutions, directly generating anchor-style videos with full-body motions remains challenging. In this study, we propose Make-Your-Anchor, a novel system necessitating only a one-minute video clip of an individual for training, subsequently enabling the automatic generation of anchor-style videos with precise torso and hand movements. Specifically, we finetune a proposed structure-guided diffusion model on input video to render 3D mesh conditions into human appearances. We adopt a two-stage training strategy for the diffusion model, effectively binding movements with specific appearances. To produce arbitrary long temporal video, we extend the 2D U-Net in the frame-wise diffusion model to a 3D style without additional training cost, and a simple yet effective batch-overlapped temporal denoising module is proposed to bypass the constraints on video length during inference. Finally, a novel identity-specific face enhancement module is introduced to improve the visual quality of facial regions in the output videos. Comparative experiments demonstrate the effectiveness and superiority of the system in terms of visual quality, temporal coherence, and identity preservation, outperforming SOTA diffusion/non-diffusion methods. Project page: \url{https://github.com/ICTMCG/Make-Your-Anchor}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_16510 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Make-Your-Anchor: A Diffusion-based 2D Avatar Generation Framework Huang, Ziyao Tang, Fan Zhang, Yong Cun, Xiaodong Cao, Juan Li, Jintao Lee, Tong-Yee Computer Vision and Pattern Recognition Despite the remarkable process of talking-head-based avatar-creating solutions, directly generating anchor-style videos with full-body motions remains challenging. In this study, we propose Make-Your-Anchor, a novel system necessitating only a one-minute video clip of an individual for training, subsequently enabling the automatic generation of anchor-style videos with precise torso and hand movements. Specifically, we finetune a proposed structure-guided diffusion model on input video to render 3D mesh conditions into human appearances. We adopt a two-stage training strategy for the diffusion model, effectively binding movements with specific appearances. To produce arbitrary long temporal video, we extend the 2D U-Net in the frame-wise diffusion model to a 3D style without additional training cost, and a simple yet effective batch-overlapped temporal denoising module is proposed to bypass the constraints on video length during inference. Finally, a novel identity-specific face enhancement module is introduced to improve the visual quality of facial regions in the output videos. Comparative experiments demonstrate the effectiveness and superiority of the system in terms of visual quality, temporal coherence, and identity preservation, outperforming SOTA diffusion/non-diffusion methods. Project page: \url{https://github.com/ICTMCG/Make-Your-Anchor}. |
| title | Make-Your-Anchor: A Diffusion-based 2D Avatar Generation Framework |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.16510 |