Skip to main navigation Skip to search Skip to main content

Enhance Radar Point Cloud with 2D Diffusion

  • Yinbao LI*
  • , Yulei ZHANG
  • , Rui ZHU
  • , Chang LIU
  • *Corresponding author for this work

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

Abstract

We propose a novel local diffusion aided single stage detector for radar to tackle the noise and sparsity issues of its point cloud in detection tasks. We utilized space occupancy maps to represent the downsampled point cloud and applied a diffusion module to denoise them. We also introduced instance-aware downsampling strategies and 3D upsampling module to enhance the model’s perception ability in 3D space. Experiments show that LDRadSSD outperformed those SOTA approaches in predicting bounding boxes for road users such as cyclists and pedestrians and figuring out the drivable space in bird’s eye view (BEV) of the scene. In particular, with IoU thresholds of 0.5/0.25/0.25, the average prediction precision (AP) of main type of road users (cars, pedestrians and cyclists) reached at competitive 58.5%, 63.1%, and 82.1%, respectively, while mean IoU of free space was 92.6%. Moreover, the prediction precision of object orientation dramatically raised to averaged 71.1%. We also demonstrate in this paper that LDRadSSD can satisfy real-time requirements in autonomous driving as its running speed reach at 18.2 to 41.7 milliseconds per frame.

Original languageEnglish
Title of host publicationNeural Information Processing : 31st International Conference, ICONIP 2024, Auckland, New Zealand, December 2–6, 2024, Proceedings, Part XI
EditorsMufti MAHMUD, Maryam DOBORJEH, Kevin WONG, Andrew Chi Sing LEUNG, Zohreh DOBORJEH, M. TANVEER
PublisherSpringer Science and Business Media Deutschland GmbH
Pages17-31
Number of pages15
ISBN (Print)9789819666058
DOIs
Publication statusPublished - 2025
Event31st International Conference on Neural Information Processing, ICONIP 2024 - Auckland, New Zealand
Duration: 2 Dec 20246 Dec 2024

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
VolumeLNCS 15296
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349
NameInternational Conference on Neural Information Processing
PublisherSpringer

Conference

Conference31st International Conference on Neural Information Processing, ICONIP 2024
Country/TerritoryNew Zealand
CityAuckland
Period2/12/246/12/24

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

Keywords

  • Diffusion model
  • Object detection
  • Radar point cloud

Fingerprint

Dive into the research topics of 'Enhance Radar Point Cloud with 2D Diffusion'. Together they form a unique fingerprint.

Cite this