AI Paper Atlas

Research map / 3D vision / Human pose estimation

Diffusion-Driven Self-Supervised Learning for Shape Reconstruction and Pose Estimation

Jingtao Sun, Yaonan Wang, Mingtao Feng, Chao Ding, Mike Zheng Shou, Ajmal Saeed Mian

TPAMI 2025 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 3 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 RK-Point layer (ablation)ORT module · REAL275IoU50 mAP: 60.5 -> 86.1Improved
add RK-Point layer (ablation)ORT module · REAL2755deg2cm mAP: 39.9 -> 55.0Improved
add RK-Point layer (ablation)ORT module · REAL275CD (1e-2): 0.07 -> 0.06No clear effect
add 3DSI-Graph layer (ablation)ORT module · REAL275CD (1e-2): 0.24 -> 0.06Improved
add 3DSI-Graph layer (ablation)ORT module · REAL2755deg2cm mAP: 52.9 -> 55.0No clear effect
add SE(3)-Group layer (ablation)ORT module · REAL27510deg5cm mAP: 70.0 -> 83.7Improved
add SE(3)-Group layer (ablation)ORT module · REAL275CD (1e-2): 0.10 -> 0.06Improved
add full SE(3) block (ablation)ORT module · REAL27510deg5cm mAP: 62.4 -> 83.7Improved
add full SE(3) block (ablation)ORT module · REAL275CD (1e-2): 0.30 -> 0.06Improved
add seed points aggregation (ablation)ORT module · REAL2755deg5cm mAP: 57.5 -> 59.8Improved
fine-tune on Wild6DOurs · Wild6D10deg5cm mAP: 48.4 -> 50.9Improved
diffusion-driven self-supervised training vs prior self-supervised SOTAOurs vs Self-DPDN · REAL2755deg2cm mAP: 39.4 -> 55.0Improved

Earlier methods this paper compares with

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