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

Domain Adaptation (DA) approaches achieved significant improvements in a widerange of machine learning and computer vision tasks (i.e., classification,detection, and segmentation). However, as far as we are aware, there are fewmethods yet to achieve domain adaptation directly on 3D point cloud data. Theunique challenge of point cloud data lies in its abundant spatial geometricinformation, and the semantics of the whole object is contributed by includingregional geometric structures. Specifically, most general-purpose DA methodsthat struggle for global feature alignment and ignore local geometricinformation are not suitable for 3D domain alignment. In this paper, we proposea novel 3D Domain Adaptation Network for point cloud data (PointDAN). PointDANjointly aligns the global and local features in multi-level. For localalignment, we propose Self-Adaptive (SA) node module with an adjusted receptivefield to model the discriminative local structures for aligning domains. Torepresent hierarchically scaled features, node-attention module is furtherintroduced to weight the relationship of SA nodes across objects and domains.For global alignment, an adversarial-training strategy is employed to learn andalign global features across domains. Since there is no common evaluationbenchmark for 3D point cloud DA scenario, we build a general benchmark (i.e.,PointDA-10) extracted from three popular 3D object/scene datasets (i.e.,ModelNet, ShapeNet and ScanNet) for cross-domain 3D objects classificationfashion. Extensive experiments on PointDA-10 illustrate the superiority of ourmodel over the state-of-the-art general-purpose DA methods.

PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representation