AI Paper Atlas

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

Progressive Classifier and Feature Extractor Adaptation for Unsupervised Domain Adaptation on Point Clouds

Zicheng Wang, Zhen Zhao, Yiming Wu, Luping Zhou, Dong Xu

ECCV 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Feature alignment for unsupervised adaptation · 5 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 progressive target-approaching (PTA) to TSFA + IDFA (ablation)PCFEA · PointDA-10average accuracy (%): 68.6 -> 76.5Improved
combine TSFA and IDFA directly without intermediate domains (ablation)PCFEA · PointDA-10average accuracy (%): 74.3 -> 68.6Hurt
add target-styled feature augmentation (TSFA) to DGCNN (ablation)DGCNN baseline · PointDA-10average accuracy (%): 62.2 -> 74.3Improved
add IDFA alone to DGCNN (ablation)DGCNN baseline · PointDA-10average accuracy (%): 62.2 -> 69.2Improved
add PTA to TSFA (ablation)PCFEA · PointDA-10average accuracy (%): 74.3 -> 75.7Improved
add PTA to IDFA (ablation)PCFEA · PointDA-10average accuracy (%): 69.2 -> 74.0Improved
add IDFA to TSFA + PTA (ablation)PCFEA · PointDA-10average accuracy (%): 75.7 -> 76.5Improved
progressively decrease selected source samples (ablation)PTA · PointDA-10average accuracy (%): 74.3 -> 75.4Improved
progressively increase selected target samples (ablation)PTA · PointDA-10average accuracy (%): 74.3 -> 74.7Mixed
shorten intermediate domain stage length tau (more frequent updates) (ablation)PCFEA · PointDA-10average accuracy (%)Improved
PCFEA vs DAPS without pseudo-label fine-tuningPCFEA · GraspNetPC-10average accuracy (%): 80.7 -> 87.6Improved

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