CameraCtrl: Enabling Camera Control for Text-to-Video Generation

By Hao He, Yinghao Xu, Yuwei Guo, Gordon Wetzstein, Bo Dai, Hongsheng Li, Ceyuan Yang. arXiv.org, 2024

425 citations recorded by the source corpus.

Controllability plays a crucial role in video generation, as it allows users to create and edit content more precisely. Existing models, however, lack control of camera pose that serves as a cinematic language to express deeper narrative nuances. To alleviate this issue, we introduce CameraCtrl, enabling accurate camera pose control for video diffusion models. Our approach explores effective camera trajectory parameterization along with a plug-and-play camera pose control module that is trained on top of a video diffusion model, leaving other modules of the base model untouched. Moreover, a comprehensive study on the effect of various training datasets is conducted, suggesting that videos with diverse camera distributions and similar appearance to the base model indeed enhance controllability and generalization. Experimental results demonstrate the effectiveness of CameraCtrl in achieving precise camera control with different video generation models, marking a step forward in the pursuit of dynamic and customized video storytelling from textual and camera pose inputs.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/3fc6184c72fd55b67409ef87493c333f15a33180

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Ying Shan
2. Huan Ling
3. Tianchang Shen
4. Jinbo Xing
5. Xiaoyu Li
6. Wangbo Yu
7. Li Yuan
8. Xiangjun Gao

In Control and Systems Engineering

1. Ying Shan

Citations in the corpus

• ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis (2024, methodology): “Following [58], we transform the camera coordinate of the estimated poses to be relative to the first frame, and normalize the translation scale using the furthest frame.”
• Gen3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control (2025, positive): “We compare G EN 3C to four baselines, including GenWarp [46], MotionCtrl [59], Camer-aCtrl [16], and NVS-Solver [66].”