Taming Camera-Controlled Video Generation with Verifiable Geometry Reward

TL;DR

This paper introduces an online reinforcement learning framework with a verifiable geometry reward to enhance camera-controlled video generation, achieving better accuracy, consistency, and quality over traditional supervised fine-tuning methods.

Contribution

It presents a novel RL post-training approach with a geometry-based reward for improved camera control in video generation, along with a new dataset for diverse camera motions.

Findings

Outperforms supervised fine-tuning in camera-control accuracy

Improves geometric consistency in generated videos

Enhances visual quality of camera-controlled videos

Abstract

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.