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

3D semantic segmentation on multi-scan large-scale point clouds plays animportant role in autonomous systems. Unlike the single-scan-based semanticsegmentation task, this task requires distinguishing the motion states ofpoints in addition to their semantic categories. However, methods designed forsingle-scan-based segmentation tasks perform poorly on the multi-scan task dueto the lacking of an effective way to integrate temporal information. Wepropose MarS3D, a plug-and-play motion-aware module for semantic segmentationon multi-scan 3D point clouds. This module can be flexibly combined withsingle-scan models to allow them to have multi-scan perception abilities. Themodel encompasses two key designs: the Cross-Frame Feature Embedding module forenriching representation learning and the Motion-Aware Feature Learning modulefor enhancing motion awareness. Extensive experiments show that MarS3D canimprove the performance of the baseline model by a large margin. The code isavailable at https://github.com/CVMI-Lab/MarS3D.

MarS3D: A Plug-and-Play Motion-Aware Model for Semantic Segmentation on Multi-Scan 3D Point Clouds