Towards Understanding Camera Motions in Any Video

Siyuan Cen, Hewei Wang, Chancharik Mitra, Daniel Jiang, Jay Karhade, Yu Tong Tiffany Ling, Jiarui Li, Yuhan Huang, Rushikesh Zawar, Xue Bai, Yilun Du, Deva Ramanan, Zhiqiu Lin

Keywords: Video-Language Models, Camera Motion, Video Understanding, Structure-from-Motion

TL;DR: We present CameraBench, which is a 3K-video benchmark advancing camera motion understanding across humans and models.

Abstract: We introduce CameraBench, a large-scale dataset and benchmark designed to assess and improve camera motion understanding. CameraBench consists of ~3,000 diverse internet videos, annotated by experts through a rigorous multi-stage quality control process. One of our core contributions is a taxonomy or "language" of camera motion primitives, designed in collaboration with cinematographers. We find, for example, that some motions like "follow" (or tracking) require understanding scene content like moving subjects. We conduct a large-scale human study to quantify human performance, revealing that domain expertise and tutorial-based training can significantly enhance accuracy. For example, a novice may confuse zoom-in (a change of intrinsics) with translating forward (a change of extrinsics), but can be trained to differentiate the two. Using CameraBench, we evaluate Structure-from-Motion (SfM) and Video-Language Models (VLMs), finding that SfM models struggle to capture semantic primitives that depend on scene content, while generative VLMs struggle to capture geometric primitives that require precise estimation of trajectories. We then fine-tune a generative VLM on CameraBench to achieve the best of both worlds and showcase its applications, including motion-augmented captioning, video question answering, and video-text retrieval. We hope our taxonomy, benchmark, and tutorials will drive future efforts towards the ultimate goal of understanding camera motions in any video.

Submission Number: 12