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

Edge computing-based 3D perception has received attention in intelligenttransportation systems (ITS) because real-time monitoring of traffic candidatespotentially strengthens Vehicle-to-Everything (V2X) orchestration. Thanks tothe capability of precisely measuring the depth information on surroundingsfrom LiDAR, the increasing studies focus on lidar-based 3D detection, whichsignificantly promotes the development of 3D perception. Few methods met thereal-time requirement of edge deployment because of high computation-intensiveoperations. Moreover, an inconsistency problem of object detection remainsuncovered in the pointcloud domain due to large sparsity. This paper thoroughlyanalyses this problem, comprehensively roused by recent works on determininginconsistency problems in the image specialisation. Therefore, we proposed a 3Dharmonic loss function to relieve the pointcloud based inconsistentpredictions. Moreover, the feasibility of 3D harmonic loss is demonstrated froma mathematical optimization perspective. The KITTI dataset and DAIR-V2X-Idataset are used for simulations, and our proposed method considerably improvesthe performance than benchmark models. Further, the simulative deployment on anedge device (Jetson Xavier TX) validates our proposed model's efficiency.

3D Harmonic Loss: Towards Task-consistent and Time-friendly 3D Object Detection on Edge for V2X Orchestration