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

Self-supervised learning on point clouds has gained a lot of attentionrecently, since it addresses the label-efficiency and domain-gap problems onpoint cloud tasks. In this paper, we propose a novel self-supervised frameworkto learn informative representations from partial point clouds. We leveragepartial point clouds scanned by LiDAR that contain both content and poseattributes, and we show that disentangling such two factors from partial pointclouds enhances feature representation learning. To this end, our frameworkconsists of three main parts: 1) a completion network to capture holisticsemantics of point clouds; 2) a pose regression network to understand theviewing angle where partial data is scanned from; 3) a partial reconstructionnetwork to encourage the model to learn content and pose features. Todemonstrate the robustness of the learnt feature representations, we conductseveral downstream tasks including classification, part segmentation, andregistration, with comparisons against state-of-the-art methods. Our method notonly outperforms existing self-supervised methods, but also shows a bettergeneralizability across synthetic and real-world datasets.

Self-Supervised Feature Learning from Partial Point Clouds via Pose Disentanglement