AI Paper Atlas

Research map / 3D vision / Human pose estimation

End2End Multi-View Feature Matching with Differentiable Pose Optimization

Barbara Roessle, Matthias Nießner

ICCV 2023 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 30 citations (OpenAlex)

Read the paperCheck 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 multi-view matching in the GNN (ablation)Ours · Matterport3D 5-viewpose AUC: average drop of 14.2% when omittedImproved
add end-to-end training with differentiable pose optimization (ablation)Ours · Matterport3D 5-viewpose AUC: average drop of 7.3% when omittedImproved
use weighted 8-point plus bundle adjustment instead of RANSAC at inference (ablation)Ours w/o multi-view · ScanNet two-viewpose error AUC@5: 20.7 -> 25.7Improved
use weighted 8-point on baseline matchesLoFTR / 3DG-STFM · MegaDepth two-viewpose error AUC@5: 3DG-STFM 52.6 -> 4.0Hurt
joint multi-view matching instead of pairwise matching on 5-tuplesOurs · ScanNet runtimematching time: 371 ms -> 338 ms (SuperGlue vs Ours)Improved
use confidence thresholding before RANSACOurs · ScanNet runtimeRANSAC time: 126 ms -> 52 msImproved
train additionally with bundle adjustment in the loop (ablation)Ours · ScanNet two-viewpose error AUC@5: 25.7 -> 26.0No clear effect
increase number of images in joint matching window (ablation)Ours · ScanNet 9-image sequencespose error AUCImproved
use Ours w/o multi-view with RANSAC instead of LoFTRLoFTR · MegaDepth two-view RANSACpose error AUC@5: 52.8 -> 49.5Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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