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

For the SLAM system in robotics and autonomous driving, the accuracy offront-end odometry and back-end loop-closure detection determine the wholeintelligent system performance. But the LiDAR-SLAM could be disturbed bycurrent scene moving objects, resulting in drift errors and even loop-closurefailure. Thus, the ability to detect and segment moving objects is essentialfor high-precision positioning and building a consistent map. In this paper, weaddress the problem of moving object segmentation from 3D LiDAR scans toimprove the odometry and loop-closure accuracy of SLAM. We propose a novel 3DSequential Moving-Object-Segmentation (3D-SeqMOS) method that can accuratelysegment the scene into moving and static objects, such as moving and staticcars. Different from the existing projected-image method, we process the raw 3Dpoint cloud and build a 3D convolution neural network for MOS task. Inaddition, to make full use of the spatio-temporal information of point cloud,we propose a point cloud residual mechanism using the spatial features ofcurrent scan and the temporal features of previous residual scans. Besides, webuild a complete SLAM framework to verify the effectiveness and accuracy of3D-SeqMOS. Experiments on SemanticKITTI dataset show that our proposed3D-SeqMOS method can effectively detect moving objects and improve the accuracyof LiDAR odometry and loop-closure detection. The test results show our3D-SeqMOS outperforms the state-of-the-art method by 12.4%. We extend theproposed method to the SemanticKITTI: Moving Object Segmentation competitionand achieve the 2nd in the leaderboard, showing its effectiveness.

3D-SeqMOS: A Novel Sequential 3D Moving Object Segmentation in Autonomous Driving