Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.05076 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912749669318656 |
|---|---|
| author | Wang, Yiming Zhang, Qihang Cai, Shengqu Wu, Tong Ackermann, Jan Kuang, Zhengfei Zheng, Yang Rajič, Frano Tang, Siyu Wetzstein, Gordon |
| author_facet | Wang, Yiming Zhang, Qihang Cai, Shengqu Wu, Tong Ackermann, Jan Kuang, Zhengfei Zheng, Yang Rajič, Frano Tang, Siyu Wetzstein, Gordon |
| contents | Emerging video diffusion models achieve high visual fidelity but fundamentally couple scene dynamics with camera motion, limiting their ability to provide precise spatial and temporal control. We introduce a 4D-controllable video diffusion framework that explicitly decouples scene dynamics from camera pose, enabling fine-grained manipulation of both scene dynamics and camera viewpoint. Our framework takes continuous world-time sequences and camera trajectories as conditioning inputs, injecting them into the video diffusion model through a 4D positional encoding in the attention layer and adaptive normalizations for feature modulation. To train this model, we curate a unique dataset in which temporal and camera variations are independently parameterized; this dataset will be made public. Experiments show that our model achieves robust real-world 4D control across diverse timing patterns and camera trajectories, while preserving high generation quality and outperforming prior work in controllability. See our website for video results: https://19reborn.github.io/Bullet4D/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_05076 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | BulletTime: Decoupled Control of Time and Camera Pose for Video Generation Wang, Yiming Zhang, Qihang Cai, Shengqu Wu, Tong Ackermann, Jan Kuang, Zhengfei Zheng, Yang Rajič, Frano Tang, Siyu Wetzstein, Gordon Computer Vision and Pattern Recognition Emerging video diffusion models achieve high visual fidelity but fundamentally couple scene dynamics with camera motion, limiting their ability to provide precise spatial and temporal control. We introduce a 4D-controllable video diffusion framework that explicitly decouples scene dynamics from camera pose, enabling fine-grained manipulation of both scene dynamics and camera viewpoint. Our framework takes continuous world-time sequences and camera trajectories as conditioning inputs, injecting them into the video diffusion model through a 4D positional encoding in the attention layer and adaptive normalizations for feature modulation. To train this model, we curate a unique dataset in which temporal and camera variations are independently parameterized; this dataset will be made public. Experiments show that our model achieves robust real-world 4D control across diverse timing patterns and camera trajectories, while preserving high generation quality and outperforming prior work in controllability. See our website for video results: https://19reborn.github.io/Bullet4D/ |
| title | BulletTime: Decoupled Control of Time and Camera Pose for Video Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.05076 |