WeakPCSOD: Overcoming the Bias of Box Annotations for Weakly Supervised Point Cloud Salient Object Detection

Authors

  • Jun Wei FNii, CUHK-Shenzhen, Shenzhen, China SSE, CUHK-Shenzhen, Shenzhen, China
  • S. Kevin Zhou School of Biomedical Engineering & Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, China Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
  • Shuguang Cui SSE, CUHK-Shenzhen, Shenzhen, China FNii, CUHK-Shenzhen, Shenzhen, China
  • Zhen Li SSE, CUHK-Shenzhen, Shenzhen, China FNii, CUHK-Shenzhen, Shenzhen, China

DOI:

https://doi.org/10.1609/aaai.v38i6.28403

Keywords:

CV: 3D Computer Vision, CV: Low Level & Physics-based Vision

Abstract

Point cloud salient object detection (PCSOD) is a newly proposed task in 3D dense segmentation. However, the acquisition of accurate 3D dense annotations comes at a high cost, severely limiting the progress of PCSOD. To address this issue, we propose the first weakly supervised PCSOD (named WeakPCSOD) model, which relies solely on cheap 3D bounding box annotations. In WeakPCSOD, we extract noise-free supervision from coarse 3D bounding boxes while mitigating shape biases inherent in box annotations. To achieve this, we introduce a novel mask-to-box (M2B) transformation and a color consistency (CC) loss. The M2B transformation, from a shape perspective, disentangles predictions from labels, enabling the extraction of noiseless supervision from labels while preserving object shapes independently of the box bias. From an appearance perspective, we further introduce the CC loss to provide dense supervision, which mitigates the non-unique predictions stemming from weak supervision and substantially reduces prediction variability. Furthermore, we employ a self-training (ST) strategy to enhance performance by utilizing high-confidence pseudo labels. Notably, the M2B transformation, CC loss, and ST strategy are seamlessly integrated into any model and incur no computational costs for inference. Extensive experiments demonstrate the effectiveness of our WeakPCSOD model, even comparable to fully supervised models utilizing dense annotations.

Published

2024-03-24

How to Cite

Wei, J., Zhou, S. K., Cui, S., & Li, Z. (2024). WeakPCSOD: Overcoming the Bias of Box Annotations for Weakly Supervised Point Cloud Salient Object Detection. Proceedings of the AAAI Conference on Artificial Intelligence, 38(6), 5894-5902. https://doi.org/10.1609/aaai.v38i6.28403

Issue

Section

AAAI Technical Track on Computer Vision V