English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Understanding point clouds captured from the real-world is challenging due toshifts in data distribution caused by varying object scales, sensor angles, andself-occlusion. Prior works have addressed this issue by combining recentlearning principles such as self-supervised learning, self-training, andadversarial training, which leads to significant computational overhead.Towardsuccinct yet powerful domain adaptation for point clouds, we revisit the uniquechallenges of point cloud data under domain shift scenarios and discover theimportance of the global geometry of source data and trends of targetpseudo-labels biased to the source label distribution. Motivated by ourobservations, we propose an adapter-guided domain adaptation method,PC-Adapter, that preserves the global shape information of the source domainusing an attention-based adapter, while learning the local characteristics ofthe target domain via another adapter equipped with graph convolution.Additionally, we propose a novel pseudo-labeling strategy resilient to theclassifier bias by adjusting confidence scores using their class-wiseconfidence distributions to consider relative confidences. Our methoddemonstrates superiority over baselines on various domain shift settings inbenchmark datasets - PointDA, GraspNetPC, and PointSegDA.

PC-Adapter: Topology-Aware Adapter for Efficient Domain Adaption on Point Clouds with Rectified Pseudo-label