Gen4P repurposes video generators as 4D perceivers for geometry and motion
Original titleGen4P: Video Generators are 4D Perceivers
- Outlet
- Takara TLDR
- Published
- Importance
- 2/5
Summary
The authors introduce Gen4P, a framework that converts pretrained video generators into feed-forward 4D perceivers for recovering geometry and motion from monocular video. It achieves state-of-the-art performance on 3D point tracking, video depth, and camera pose estimation.
Original excerpt
We introduce Gen4P, a unified framework for recovering 4D geometry and motion from monocular video. We show that a pretrained video generator can be repurposed into a feed-forward 4D perceiver without iterative denoising. Building on on-demand querying, Gen4P processes clean video latents in a single encoder pass and predicts the requested 4D attributes through a lightweight cross-attention decoder. The on-demand query formulation enables flexible spatial sampling, and our proposed continuous-time queries further provide temporal flexibility. To strengthen geometric knowledge, we propose a straightforward plug-in solution by incorporating a pretrained video-depth LoRA as a frozen geometry prior, improving overall performance without adding extra trainable parameters. Gen4P achieves state-of-the-art or competitive performance on tasks of 3D point tracking, video depth and camera pose estimation. Qualitative results on in-the-wild videos illustrate its capability of tracking complex motion while maintaining the delicate structure. Together, these results demonstrate the potential of video generative pretraining for unified 4D perception. See https://gen4p.github.io for more visual results.