AI Paper Atlas

Research map / 3D vision / Human pose estimation

DISP6D: Disentangled Implicit Shape and Pose Learning for Scalable 6D Pose Estimation

Yilin Wen, Xiangyu Li, Hao Pan, Lei Yang, Zheng Wang, Taku Komura, Wenping Wang

ECCV 2022 · 3D vision › Human pose estimation · approach: Instance-level 6D object pose · 19 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
condition pose codebook on shape code with tensor W instead of MLP on pose encoding only (ablation)DISP6D · REAL275 setting IIIAP30: 47.0 -> 75.3Improved
use 3rd-order tensor W instead of MLP on concatenated codes (ablation)DISP6D · REAL275 setting IIIAP30: 74.9 -> 75.3Improved
use 3rd-order tensor W instead of MLP on concatenated codes (ablation)DISP6D · REAL275 setting IIIAP10: 27.3 -> 30.9Improved
stop gradient on shape code z_o in pose code generation (ablation)DISP6D · REAL275 setting IIIAP10: 26.6 -> 30.9Improved
stop gradient on shape code z_o in pose code generation (ablation)DISP6D · REAL275 setting IIIAP60: 87.5 -> 84.3Hurt
add shape contrastive loss L_shape (ablation)DISP6D · REAL275 setting IIIAP30: 67.6 -> 75.3Improved
add shape contrastive loss L_shape (ablation)DISP6D · REAL275 setting IIIAP10: 15.2 -> 30.9Improved
condition decoder on shape code z_o instead of pose code z_p (ablation)DISP6D · REAL275 setting IIIAP30: 59.9 -> 75.3Improved
use DISP6D instead of PoseContrast for cross-category rotationOurs-all · REAL275 (GT mask)rotation accuracy at 30 deg: 55.0 -> 89.9Improved
use DISP6D instead of Multipath-AAE in full detection pipelineDISP6D · T-LESS (Mask R-CNN)VSD recall Obj 1-30: 23.51 -> 35.36Improved
use DISP6D instead of Multipath-AAE with GT maskDISP6D · T-LESS (GT boxes)VSD recall Obj 19-30: 59.89 -> 64.42Improved
use DISP6D instead of per-object AAE on novel objectsDISP6D · T-LESS (GT boxes)VSD recall Obj 19-30: 66.63 -> 64.42Hurt

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