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

Real scans always miss partial geometries of objects due to theself-occlusions, external-occlusions, and limited sensor resolutions. Pointcloud completion aims to refer the complete shapes for incomplete 3D scans ofobjects. Current deep learning-based approaches rely on large-scale completeshapes in the training process, which are usually obtained from syntheticdatasets. It is not applicable for real-world scans due to the domain gap. Inthis paper, we propose a self-supervised point cloud completion method (TraPCC)for vehicles in real traffic scenes without any complete data. Based on thesymmetry and similarity of vehicles, we make use of consecutive point cloudframes to construct vehicle memory bank as reference. We design a bottom-upmechanism to focus on both local geometry details and global shape features ofinputs. In addition, we design a scene-graph in the network to pay attention tothe missing parts by the aid of neighboring vehicles. Experiments show thatTraPCC achieve good performance for real-scan completion on KITTI and nuScenestraffic datasets even without any complete data in training. We also show adownstream application of 3D detection, which benefits from our completionapproach.

Self-supervised Point Cloud Completion on Real Traffic Scenes via Scene-concerned Bottom-up Mechanism