AI Paper Atlas

Research map / 3D vision / Human pose estimation

Pose Guided RGBD Feature Learning for 3D Object Pose Estimation

Vassileios Balntas, Andreas Doumanoglou, Caner Sahin, Juil Sock, Rigas Kouskouridas, Tae-Kyun Kim

ICCV 2017 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 83 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 pose-guided term Lpose with Lobject in place of Ltriplets + Lpairsfeature learning on LINEMOD · LINEMOD (synthetic training, average)pose accuracy (%): 30.34 -> 54.32Improved
add pose-guided term without bootstrapping versus Wohlhart without bootstrappingfeature learning · LINEMOD (synthetic training, average)pose accuracy (%): 18.61 -> 51.02Improved
add symmetry weight phi to pose loss (ablation)Lobject + Lpose · LINEMOD (synthetic training, average)pose accuracy (%): 51.02 -> 54.32Improved
add regression loss Lreg (ablation)Lobject + Lpose phi · LINEMOD (synthetic training, average)pose accuracy (%): 54.32 -> 55.64Improved
add regression loss Lreg for unseen objects (protocol 3) (ablation)Lobject + Lpose · LINEMOD real data, unseen objectsmean angle errorHurt
add regression loss with dissimilar train/test poses (protocol 2) (ablation)Lobject + Lpose · LINEMOD real data, dissimilar posesmean angle errorNo clear effect
add regression loss with random split (protocol 1) (ablation)Lobject + Lpose · LINEMOD real data, random splitmean angle errorImproved
add occlusion term Locc (belt object) (ablation)pose-guided features · Hand-Object occlusion, belt objectaverage angle error: 14.3 -> 11.8Improved
add occlusion term Locc (10 objects) (ablation)pose-guided features · Hand-Object occlusion, 10 objectsaverage angle error: 36.3 -> 27.7Improved
replace Wohlhart triplet/pair loss with pose-guided loss, single NNpose descriptor · LINEMOD 5 degree accuracypose accuracy (%): 54% to 72%Improved

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