AI Paper Atlas

Research map / 3D vision / Depth and stereo

Learning To Segment Rigid Motions From Two Frames

Gengshan Yang, Deva Ramanan

CVPR 2021 · 3D vision › Depth and stereo · approach: Learned optical and scene flow · 59 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add rigidity cost-map inputs (ablation)Rigid motion segmentation network · Sintelbackground IoU (%): 76.81 to 84.63Improved
Add uncertainty inputs (ablation)Rigid motion segmentation network · KITTI-SFobject IoU (%): 85.09 to 89.53Improved
Add monocular depth inputs (ablation)Rigid motion segmentation network · KITTI-SFobject IoU (%): 84.46 to 89.53Improved
Add optical expansion to upgrade 2D flow to 3D (ablation)Rigid motion segmentation network · KITTI-SFobject IoU (%): 81.28 to 89.53Improved
Use a learned segmentation network instead of manually thresholded rigidity cost maps (ablation)Rigid motion segmentation · KITTI-SFobject IoU (%): 25.83 to 89.53Improved
Use rigid motion masks instead of Mask R-CNN masks for rigid-body scene-flow fittingScene-flow estimation · KITTI benchmarkscene flow error (%): 5.61 to 4.89Improved
Apply rigid-body refinement to optical-expansion scene flow (ablation)Scene-flow estimation · KITTI-SFscene flow error reduction (%): 24.6%Improved
Use rigid motion masks rather than appearance-based instance masksMonocular scene-flow estimation · KITTIscene flow error (%): 22.47 to 22.19Improved
Use rigid motion masks rather than appearance-based instance masksMonocular scene-flow estimation · Sintelscene flow error (%): 47.40 to 46.05Improved
Use rigid motion masks rather than appearance-based instance masks (ablation)Stereo scene-flow estimation · KITTI benchmarkscene flow error (%): 5.61 to 4.89Improved
Use rigid motion segmentation instead of optical expansion without rigid-body refinementMonocular scene-flow estimation · KITTIscene flow error (%): 44.87 to 22.19Improved
Use rigid motion segmentation instead of optical expansion without rigid-body refinementMonocular scene-flow estimation · Sintelscene flow error (%): 55.43 to 46.05Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy