Skip to main navigation Skip to search Skip to main content

PointOBB: Learning Oriented Object Detection via Single Point Supervision

  • Junwei Luo
  • , Xue Yang
  • , Yi Yu
  • , Qingyun Li
  • , Junchi Yan
  • , Yansheng Li*
  • *Corresponding author for this work
  • Shanghai AI Laboratory
  • Southeast University
  • Harbin Institute of Technology
  • Shanghai Jiao Tong University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

72 Scopus citations

Abstract

Single point-supervised object detection is gaining attention due to its cost-effectiveness. However, existing approaches focus on generating horizontal bounding boxes (HBBs) while ignoring oriented bounding boxes (OBBs) commonly used for objects in aerial images. This paper proposes PointOBB, the first single Point-based OBB generation method, for oriented object detection. PointOBB operates through the collaborative utilization of three distinctive views: an original view, a resized view, and a rotated/flipped (rot/flp) view. Upon the original view, we leverage the resized and rot/flp views to build a scale augmentation module and an angle acquisition module, respectively. In the former module, a Scale-Sensitive Consistency (SSC) loss is designed to enhance the deep network's ability to perceive the object scale. For accurate object angle predictions, the latter module incorporates self-supervised learning to predict angles, which is associated with a scale-guided Dense-to-Sparse (DS) matching strategy for aggregating dense angles corresponding to sparse objects. The resized and rot/flp views are switched using a progressive multi-view switching strategy during training to achieve coupled optimization of scale and angle. Experimental results on the DIOR-R and DOTA-v1.0 datasets demonstrate that PointOBB achieves promising performance, and significantly outperforms potential point-supervised baselines.
Original languageEnglish
Title of host publication2024 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
PublisherIEEE Computer Society
Pages16730-16740
Number of pages11
ISBN (Electronic)979-8-3503-5300-6
ISBN (Print)0769526462, 0818684976, 1424411807, 9781457703942, 9781467388511, 9781665445092, 9781665469463, 9781728171685, 9798350301298, 979-835035300-6
DOIs
StatePublished - 16 Jun 2024

Publication series

NameIEEE Conference on Computer Vision and Pattern Recognition
ISSN (Print)1063-6919

Funding

This work was supported by the National Natural Science Foundation of China under Grants (42371321, 62306069, 62222607, 72342023), National Key R&D Program of China (2022ZD0160100). The numerical calculations in this paper have been done on the supercomputing system in the Supercomputing Center of Wuhan University.

Indexed by

  • Scopus
  • EI
  • CPCI-S

Fingerprint

Dive into the research topics of 'PointOBB: Learning Oriented Object Detection via Single Point Supervision'. Together they form a unique fingerprint.

Cite this