AI Paper Atlas

Research map / 3D vision / Human pose estimation

Towards Human-Level 3D Relative Pose Estimation: Generalizable, Training-Free, With Single Reference

Yuan Gao, Yajing Luo, Junhong Wang, Kui Jia, Gui-Song Xia

TPAMI 2025 · 3D vision › Human pose estimation · approach: Category-level object pose estimation · 1 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 back-surface culling (ablation)render-and-compare pose refinement · LineMODMean Err: 38.09 -> 29.93Improved
add semantic maps to RGB (ablation)render-and-compare pose refinement · LineMODAcc@5: 19.62 -> 24.32Improved
add RGB maps to semantic maps (ablation)render-and-compare pose refinement · LineMODAcc@5: 19.22 -> 24.32Improved
use DINOv2 instead of LoFTR features (ablation)render-and-compare pose refinement · LineMODAcc@5: 5.24 -> 19.22Improved
use DINOv2 instead of Stable Diffusion features (ablation)render-and-compare pose refinement · LineMODAcc@5: 12.12 -> 19.22Improved
add LoFTR features to RGB (ablation)render-and-compare pose refinement · LineMODAcc@5: 19.62 -> 14.80Hurt
add SD features to RGB (ablation)render-and-compare pose refinement · LineMODMean Err: 36.26 -> 33.78Improved
use uniform instead of random initialization (ablation)render-and-compare pose refinement · LineMODAcc@5: 23.08 -> 24.32Improved
add differentiable-renderer refinement after initialization (ablation)Ours · LineMODAcc@5: 6.66 -> 24.32Improved
increase uniform initialization samples beyond 16000 (ablation)Ours · LineMODAcc@10: 4000 to 16000 marginal; saturates after 16000No clear effect
increase refinement iterations beyond 120 (ablation)Ours · LineMODAcc@10: 30 to 120 marginal; saturates after 120No clear effect
training-free method versus supervised DVMNet (in-dataset)Ours · LineMODAcc@5: 2.72 -> 24.32Improved

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