Research map / 3D vision / 3D generation
Track3R: Joint Point Map and Trajectory Prior for Spatiotemporal 3D Understanding
Seong Hyeon Park, Jinwoo Shin
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Track3R jointly models point maps and object trajectories as a prior to improve spatiotemporal 3D understanding from monocular video.
Problem
Understanding 3D scenes and dynamic objects from casual 2D monocular videos is underdetermined; existing end-to-end point map models struggle when objects are affected by motion or dynamics.
Approach
Combines pre-trained point map model priors with a trajectory prior so that object geometry and motion are estimated jointly across frames, rather than relying on point maps alone.
Closest papers by meaning
- PointZero: 3D Point Track Completion for Learning Transferable 3D Dynamics (NeurIPS 2026 · similarity 0.90)
- TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Tracking (NeurIPS 2026 · similarity 0.90)
- Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos (CVPR 2025 · similarity 0.89)
- MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion (ICLR 2025 · similarity 0.88)
- Tracktention: Leveraging Point Tracking to Attend Videos Faster and Better (CVPR 2025 · similarity 0.88)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
