Research map / Video / Video segmentation and tracking
Learning to Correct Geometry in Generated Videos
Read the paperCheck your idea against 108,967 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.
The paper argues that geometric inconsistency in generated videos under camera motion is largely a data problem, since web video supplies sparse camera-motion supervision.
Problem
Generated videos lose geometric consistency under camera motion (structures warp, parallax becomes inconsistent, smooth camera moves collapse into shot-like transitions), which the authors attribute largely to scarce camera-motion supervision in web-scale video corpora.
Approach
Frames the failure as a data problem: web videos are dominated by dynamic objects with static or weakly moving cameras, so camera-motion supervision is sparse; the paper addresses this with a learning approach to correct geometry (details not stated in the abstract).
Closest papers by meaning
- Geometry-as-context: Modulating Explicit 3D in Scene-consistent Video Generation to Geometry Context (CVPR 2026 · similarity 0.89)
- VideoGPA: Distilling Geometry Priors for 3D-Consistent Video Generation (ICML 2026 · similarity 0.89)
- GeoVideo: Introducing Geometric Regularization into Video Generation Model (NeurIPS 2025 · similarity 0.87)
- VideoMDM: Towards 3D Human Motion Generation From 2D Supervision (NeurIPS 2026 · similarity 0.87)
- PanoWorld: Geometry-Consistent Panoramic Video World Modeling (NeurIPS 2026 · similarity 0.87)
- DynaTokens: Teaching Dynamics to Camera-Controlled Video Models at Test Time (NeurIPS 2026 · similarity 0.87)
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.