Taming Camera-Controlled Video Generation with Verifiable Geometry Reward
This work addresses the challenge of generating videos with accurate camera motions for applications in content creation and simulation, representing an incremental advance in video generation methods.
The paper tackled the problem of imprecise camera control in video generation by introducing an online reinforcement learning post-training framework with a verifiable geometry reward, achieving clear improvements over supervised fine-tuning baselines in camera-control accuracy, geometric consistency, and visual quality.
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.