AI Paper Atlas

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

Domain Adaptive Hand Keypoint and Pixel Localization in the Wild

Takehiko Ohkawa, Yu-Jhe Li, Qichen Fu, Ryosuke Furuta, Kris M. Kitani, Yoichi Sato

ECCV 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 21 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 multi-task adaptation (joint keypoint and mask) instead of single-task (ablation)C-GAC · DexYCB to HO3DIoU: improved by 2.7 / 2.6 over the single-task adaption in hand segmentationImproved
add multi-task adaptation instead of single-task, keypoint branch (ablation)C-GAC · DexYCB to HO3DPCK: 50.1/51.0 in the PCK single-task; marginal gainNo clear effect
add confidence weighting from two-teacher divergence to GAC-Distill (ablation)GAC-Distill · DexYCB to HO3D, HanCo, FPHAaverage PCK/IoUImproved
replace EMA teacher update with distillation-based teacher update (ablation)GAC + MT · DexYCB to HO3Dhand segmentation IoUImproved
add geometric augmentation consistency (GAC) self-trainingSource only · DexYCB to HanCoaverage score: up to 47.4/47.9 in the average scoreImproved
use adversarial adaptation (DANN, RegDA) instead of consistency trainingGAC · DexYCB to HO3D, HanCo, FPHAaverage scoreHurt
add UMA dropout-based confidence to GACGAC · DexYCB to HO3D, HanCo, FPHAaverage scoreMixed
add CPL confident instance selection to GACGAC · DexYCB to HO3D, HanCo, FPHAkeypoint regressionMixed
add C-GAC adaptation versus source onlySource only · DexYCB to HO3Daverage score: improved by 5.3/14.4 in the average task scoreImproved

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