Moaw: Unleashing Motion Awareness for Video Diffusion Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917209698205696 |
|---|---|
| author | Zhang, Tianqi Wang, Ziyi Zheng, Wenzhao Chen, Weiliang Huang, Yuanhui Huang, Zhengyang Zhou, Jie Lu, Jiwen |
| author_facet | Zhang, Tianqi Wang, Ziyi Zheng, Wenzhao Chen, Weiliang Huang, Yuanhui Huang, Zhengyang Zhou, Jie Lu, Jiwen |
| contents | Video diffusion models, trained on large-scale datasets, naturally capture correspondences of shared features across frames. Recent works have exploited this property for tasks such as optical flow prediction and tracking in a zero-shot setting. Motivated by these findings, we investigate whether supervised training can more fully harness the tracking capability of video diffusion models. To this end, we propose Moaw, a framework that unleashes motion awareness for video diffusion models and leverages it to facilitate motion transfer. Specifically, we train a diffusion model for motion perception, shifting its modality from image-to-video generation to video-to-dense-tracking. We then construct a motion-labeled dataset to identify features that encode the strongest motion information, and inject them into a structurally identical video generation model. Owing to the homogeneity between the two networks, these features can be naturally adapted in a zero-shot manner, enabling motion transfer without additional adapters. Our work provides a new paradigm for bridging generative modeling and motion understanding, paving the way for more unified and controllable video learning frameworks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12761 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Moaw: Unleashing Motion Awareness for Video Diffusion Models Zhang, Tianqi Wang, Ziyi Zheng, Wenzhao Chen, Weiliang Huang, Yuanhui Huang, Zhengyang Zhou, Jie Lu, Jiwen Computer Vision and Pattern Recognition Video diffusion models, trained on large-scale datasets, naturally capture correspondences of shared features across frames. Recent works have exploited this property for tasks such as optical flow prediction and tracking in a zero-shot setting. Motivated by these findings, we investigate whether supervised training can more fully harness the tracking capability of video diffusion models. To this end, we propose Moaw, a framework that unleashes motion awareness for video diffusion models and leverages it to facilitate motion transfer. Specifically, we train a diffusion model for motion perception, shifting its modality from image-to-video generation to video-to-dense-tracking. We then construct a motion-labeled dataset to identify features that encode the strongest motion information, and inject them into a structurally identical video generation model. Owing to the homogeneity between the two networks, these features can be naturally adapted in a zero-shot manner, enabling motion transfer without additional adapters. Our work provides a new paradigm for bridging generative modeling and motion understanding, paving the way for more unified and controllable video learning frameworks. |
| title | Moaw: Unleashing Motion Awareness for Video Diffusion Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.12761 |