AI Paper Atlas

Research map / 3D vision / Human pose estimation

RNNPose: Recurrent 6-DoF Object Pose Refinement With Robust Correspondence Field Estimation and Pose Optimization

Yan Xu, Kwan-Yee Lin, Guofeng Zhang, Xiaogang Wang, Hongsheng Li

CVPR 2022 · 3D vision › Human pose estimation · approach: Instance-level 6D object pose · 86 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
Use recurrent correspondence-based pose optimization with differentiable LMRNNPose · LINEMODADD(-S) accuracy: 97.37 vs 96.1Improved
Use recurrent correspondence-based pose optimization with descriptor weightingRNNPose · Occlusion LINEMODADD(-S) accuracy: 51.6 -> 60.65Improved
Use RNNPose refinement after PoseCNN initializationYCB-Video · YCB-VideoAUC, ADD(-S): 61.3 -> 83.1Improved
Add correspondence-field supervision (ablation)RNNPose · LINEMODADD(-S) accuracy at 0.02d: 21.85 -> 50.39Improved
Add model alignment loss and end-to-end pose supervision (ablation)RNNPose · LINEMODADD(-S) accuracy at 0.02d: 33.72 -> 50.39Improved
Add rigid-pose correspondence-field rectification (ablation)RNNPose · LINEMODADD(-S) accuracy at 0.02d: 44.68 -> 50.39Improved
Add 3D context features instead of omitting context (ablation)RNNPose · LINEMODADD(-S) accuracy at 0.02d: 44.82 -> 50.39Improved
Use 3D context features instead of 2D context features (ablation)RNNPose · LINEMODADD(-S) accuracy at 0.02d: 47.44 -> 50.39Improved
Add descriptor-based similarity score weighting for unreliable correspondences (ablation)RNNPose · Occlusion LINEMODADD(-S) accuracy at 0.05d: 27.21 -> 31.15Improved

Later papers that compared with this method

1 later accepted paper reports RNNPose as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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