Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.21299 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910062960705536 |
|---|---|
| author | Hu, Bin Qi, Zipeng Huang, Guoxi Xu, Zunnan Zhang, Ruicheng Ye, Chongjie Zhou, Jun Li, Xiu Wang, Jingdong |
| author_facet | Hu, Bin Qi, Zipeng Huang, Guoxi Xu, Zunnan Zhang, Ruicheng Ye, Chongjie Zhou, Jun Li, Xiu Wang, Jingdong |
| contents | Single-view reference-to-video methods often struggle to preserve identity consistency under large facial-angle variations. This limitation naturally motivates the incorporation of multi-view facial references. However, simply introducing additional reference images exacerbates the \textit{copy-paste} problem, particularly the \textbf{\textit{view-dependent copy-paste}} artifact, which reduces facial motion naturalness. Although cross-paired data can alleviate this issue, collecting such data is costly. To balance the consistency and naturalness, we propose $\mathrm{Mv}^2\mathrm{ID}$, a multi-view conditioned framework under in-paired supervision. We introduce a region-masking training strategy to prevent shortcut learning and extract essential identity features by encouraging the model to aggregate complementary identity cues across views. In addition, we design a reference decoupled-RoPE mechanism that assigns distinct positional encoding to video and conditioning tokens for better modeling of their heterogeneous properties. Furthermore, we construct a large-scale dataset with diverse facial-angle variations and propose dedicated evaluation metrics for identity consistency and motion naturalness. Extensive experiments demonstrate that our method significantly improves identity consistency while maintaining motion naturalness, outperforming existing approaches trained with cross-paired data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_21299 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Identity-Consistent Video Generation under Large Facial-Angle Variations Hu, Bin Qi, Zipeng Huang, Guoxi Xu, Zunnan Zhang, Ruicheng Ye, Chongjie Zhou, Jun Li, Xiu Wang, Jingdong Computer Vision and Pattern Recognition Single-view reference-to-video methods often struggle to preserve identity consistency under large facial-angle variations. This limitation naturally motivates the incorporation of multi-view facial references. However, simply introducing additional reference images exacerbates the \textit{copy-paste} problem, particularly the \textbf{\textit{view-dependent copy-paste}} artifact, which reduces facial motion naturalness. Although cross-paired data can alleviate this issue, collecting such data is costly. To balance the consistency and naturalness, we propose $\mathrm{Mv}^2\mathrm{ID}$, a multi-view conditioned framework under in-paired supervision. We introduce a region-masking training strategy to prevent shortcut learning and extract essential identity features by encouraging the model to aggregate complementary identity cues across views. In addition, we design a reference decoupled-RoPE mechanism that assigns distinct positional encoding to video and conditioning tokens for better modeling of their heterogeneous properties. Furthermore, we construct a large-scale dataset with diverse facial-angle variations and propose dedicated evaluation metrics for identity consistency and motion naturalness. Extensive experiments demonstrate that our method significantly improves identity consistency while maintaining motion naturalness, outperforming existing approaches trained with cross-paired data. |
| title | Identity-Consistent Video Generation under Large Facial-Angle Variations |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.21299 |