SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds

Xinge Zhu*, Yuexin Ma, Tai Wamg, Yan Xu, Jianping Shi, Dahua Lin

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

38 Citations (Scopus)


Multi-class 3D object detection aims to localize and classify objects of multiple categories from point clouds. Due to the nature of point clouds, i.e. unstructured, sparse and noisy, some features benefitting multi-class discrimination are underexploited, such as shape information. In this paper, we propose a novel 3D shape signature to explore the shape information from point clouds. By incorporating operations of symmetry, convex hull and Chebyshev fitting, the proposed shape signature is not only compact and effective but also robust to the noise, which serves as a soft constraint to improve the feature capability of multi-class discrimination. Based on the proposed shape signature, we develop the shape signature networks (SSN) for 3D object detection, which consist of pyramid feature encoding part, shape-aware grouping heads and explicit shape encoding objective. Experiments show that the proposed method performs remarkably better than existing methods on two large-scale datasets. Furthermore, our shape signature can act as a plug-and-play component and ablation study shows its effectiveness and good scalability (Source code at SSN and also available at mmdetection3d soon.).
Original languageEnglish
Title of host publicationComputer Vision – ECCV 2020
Subtitle of host publication16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXV
EditorsAndrea Vedaldi, Horst Bischof, Thomas Brox, Jan-Michael Frahm
PublisherSpringer Cham
Number of pages17
ISBN (Electronic)9783030585952
ISBN (Print)9783030585945
Publication statusPublished - 28 Nov 2020
Externally publishedYes
Event16th European Conference on Computer Vision, ECCV 2020 - Glasgow, United Kingdom
Duration: 23 Aug 202028 Aug 2020

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349
NameImage Processing, Computer Vision, Pattern Recognition, and Graphics (LNIP)
ISSN (Print)3004-9946
ISSN (Electronic)3004-9954
NameECCV: European Conference on Computer Vision


Conference16th European Conference on Computer Vision, ECCV 2020
Country/TerritoryUnited Kingdom

Scopus Subject Areas

  • Theoretical Computer Science
  • Computer Science(all)


Dive into the research topics of 'SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds'. Together they form a unique fingerprint.

Cite this