You Only Need One Thing One Click: Self-Training for Weakly Supervised 3D Scene Understanding

Author:

Liu Zhengzhe1ORCID,Qi Xiaojuan2ORCID,Fu Chi-Wing1ORCID

Affiliation:

1. Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong

2. Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong

Abstract

Understanding 3D scenes, such as semantic segmentation and instance identification within point clouds, typically demands extensive annotated datasets. However, generating point-by-point labels is an overly laborious process. While recent techniques have been developed to train 3D networks with a minimal fraction of labeled points, our method, dubbed “One Thing One Click,” simplifies this by requiring just a single label per object. To effectively utilize these sparse annotations during network training, we’ve crafted an innovative self-training strategy. This involves alternating between training phases and label spreading, powered by a graph propagation module. Additionally, we integrate a relation network to create category-specific prototypes, improving pseudo label accuracy and steering the training process. Our approach also seamlessly integrates with 3D instance segmentation, incorporating a point-clustering technique. Our method demonstrates superior performance over other weakly supervised strategies for 3D semantic and instance segmentation, as evidenced by tests on both ScanNet-v2 and S3DIS datasets. Remarkably, the efficacy of our self-training method with limited annotations rivals that of fully supervised models. Codes and models are available at https://github.com/liuzhengzhe/One-Thing-One-Click .

Publisher

World Scientific Pub Co Pte Ltd

Reference72 articles.

1. J. Wei, G. Lin, K.H. Yap, T.Y. Hung and L. Xie, Proc IEEE/CVF Conf Computer Vision and Pattern Recognition (CVPR), IEEE, 2020, pp. 4384–4393.

2. X. Xu and G. H. Lee, Proc IEEE/CVF Conf Computer Vision and Pattern Recognition (CVPR), IEEE, 2020, pp. 13706–13715.

3. M. Li, Y. Xie, Y. Shen, B. Ke, R. Qiao, B. Ren, S. Lin and L. Ma, Proc IEEE/CVF Conf Computer Vision and Pattern Recognition, IEEE, 2022, pp. 14930–14939.

4. SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Clouds

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3