Mengtian Li, Yuan Xie, Yunhang Shen, Bo Ke, Ruizhi Qiao, Bohan Ren, Shaohui Lin, Lizhuang Ma
{"title":"HybridCR: Weakly-Supervised 3D Point Cloud Semantic Segmentation via Hybrid Contrastive Regularization","authors":"Mengtian Li, Yuan Xie, Yunhang Shen, Bo Ke, Ruizhi Qiao, Bohan Ren, Shaohui Lin, Lizhuang Ma","doi":"10.1109/CVPR52688.2022.01451","DOIUrl":null,"url":null,"abstract":"To address the huge labeling cost in large-scale point cloud semantic segmentation, we propose a novel hybrid contrastive regularization (HybridCR) framework in weakly-supervised setting, which obtains competitive performance compared to its fully-supervised counterpart. Specifically, HybridCR is the first framework to leverage both point consistency and employ contrastive regularization with pseudo labeling in an end-to-end manner. Fundamentally, HybridCR explicitly and effectively considers the semantic similarity between local neighboring points and global characteristics of 3D classes. We further design a dynamic point cloud augmentor to generate diversity and robust sample views, whose transformation parameter is jointly optimized with model training. Through extensive experiments, HybridCR achieves significant performance improvement against the SOTA methods on both indoor and outdoor datasets, e.g., S3DIS, ScanNet-V2, Semantic3D, and SemanticKITTI.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"28","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CVPR52688.2022.01451","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 28
Abstract
To address the huge labeling cost in large-scale point cloud semantic segmentation, we propose a novel hybrid contrastive regularization (HybridCR) framework in weakly-supervised setting, which obtains competitive performance compared to its fully-supervised counterpart. Specifically, HybridCR is the first framework to leverage both point consistency and employ contrastive regularization with pseudo labeling in an end-to-end manner. Fundamentally, HybridCR explicitly and effectively considers the semantic similarity between local neighboring points and global characteristics of 3D classes. We further design a dynamic point cloud augmentor to generate diversity and robust sample views, whose transformation parameter is jointly optimized with model training. Through extensive experiments, HybridCR achieves significant performance improvement against the SOTA methods on both indoor and outdoor datasets, e.g., S3DIS, ScanNet-V2, Semantic3D, and SemanticKITTI.