AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Source-free Domain Adaptive Human Pose Estimation

Qucheng Peng, Ce Zheng, Chen Chen

ICCV 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 38 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 the target-relevant (TR) module to MMT (ablation)MMT · RHD -> H3DPCK@0.05: 75.3 -> 79.1Improved
add the source-protect (SP) module in place of MMT (ablation)MMT + TR · RHD -> H3DPCK@0.05: 79.1 -> 82.2Improved
add pose-specific contrastive loss L_cst (ablation)baseline (L_ft + L_con) · RHD -> H3DPCK@0.05: 79.4 -> 80.9Improved
add residual loss L_res (ablation)baseline (L_ft + L_con) · SURREAL -> Human3.6MPCK@0.05: 75.8 -> 76.6Improved
add pose-specific information maximization L_im (ablation)baseline (L_ft + L_con) · SURREAL -> Human3.6MPCK@0.05: 75.8 -> 76.9Improved
use heatmap-based contrastive learning instead of no contrastive term (ablation)MMT · RHD -> H3DPCK@0.05: 75.3 -> 74.2Hurt
use vector-based spatial space for contrastive learning (ablation)MMT · RHD -> H3DPCK@0.05: 75.3 -> 78.7Improved
use vector-based instead of heatmap-based information maximization (ablation)MMT · RHD -> H3DPCK@0.05: 75.9 -> 77.5Improved
vary alpha, beta, gamma hyperparameters (ablation)Ours · RHD -> H3DPCK@0.05: decreases limited to 0.2%No clear effect
compare source-free method against UniFrame that uses source dataOurs · SURREAL -> Human3.6MPCK@0.05: 79.0 -> 78.7Hurt

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