Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.02870 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912743636860928 |
|---|---|
| author | Wang, Zhaoqing Xia, Xiaobo Bie, Zhuolin Liu, Jinlin Yu, Dongdong Bian, Jia-Wang Wang, Changhu |
| author_facet | Wang, Zhaoqing Xia, Xiaobo Bie, Zhuolin Liu, Jinlin Yu, Dongdong Bian, Jia-Wang Wang, Changhu |
| contents | Recent advances in video diffusion models have remarkably improved camera-controlled video generation, but most methods rely solely on supervised fine-tuning (SFT), leaving online reinforcement learning (RL) post-training largely underexplored. In this work, we introduce an online RL post-training framework that optimizes a pretrained video generator for precise camera control. To make RL effective in this setting, we design a verifiable geometry reward that delivers dense segment-level feedback to guide model optimization. Specifically, we estimate the 3D camera trajectories for both generated and reference videos, divide each trajectory into short segments, and compute segment-wise relative poses. The reward function then compares each generated-reference segment pair and assigns an alignment score as the reward signal, which helps alleviate reward sparsity and improve optimization efficiency. Moreover, we construct a comprehensive dataset featuring diverse large-amplitude camera motions and scenes with varied subject dynamics. Extensive experiments show that our online RL post-training clearly outperforms SFT baselines across multiple aspects, including camera-control accuracy, geometric consistency, and visual quality, demonstrating its superiority in advancing camera-controlled video generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_02870 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Taming Camera-Controlled Video Generation with Verifiable Geometry Reward Wang, Zhaoqing Xia, Xiaobo Bie, Zhuolin Liu, Jinlin Yu, Dongdong Bian, Jia-Wang Wang, Changhu Computer Vision and Pattern Recognition Recent advances in video diffusion models have remarkably improved camera-controlled video generation, but most methods rely solely on supervised fine-tuning (SFT), leaving online reinforcement learning (RL) post-training largely underexplored. In this work, we introduce an online RL post-training framework that optimizes a pretrained video generator for precise camera control. To make RL effective in this setting, we design a verifiable geometry reward that delivers dense segment-level feedback to guide model optimization. Specifically, we estimate the 3D camera trajectories for both generated and reference videos, divide each trajectory into short segments, and compute segment-wise relative poses. The reward function then compares each generated-reference segment pair and assigns an alignment score as the reward signal, which helps alleviate reward sparsity and improve optimization efficiency. Moreover, we construct a comprehensive dataset featuring diverse large-amplitude camera motions and scenes with varied subject dynamics. Extensive experiments show that our online RL post-training clearly outperforms SFT baselines across multiple aspects, including camera-control accuracy, geometric consistency, and visual quality, demonstrating its superiority in advancing camera-controlled video generation. |
| title | Taming Camera-Controlled Video Generation with Verifiable Geometry Reward |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.02870 |