TSG-Seg: Temporal-selective guidance for semi-supervised semantic segmentation of 3D LiDAR point clouds

IF 10.6 1区 地球科学 Q1 GEOGRAPHY, PHYSICAL ISPRS Journal of Photogrammetry and Remote Sensing Pub Date : 2024-08-08 DOI:10.1016/j.isprsjprs.2024.07.020
Weihao Xuan , Heli Qi , Aoran Xiao
{"title":"TSG-Seg: Temporal-selective guidance for semi-supervised semantic segmentation of 3D LiDAR point clouds","authors":"Weihao Xuan ,&nbsp;Heli Qi ,&nbsp;Aoran Xiao","doi":"10.1016/j.isprsjprs.2024.07.020","DOIUrl":null,"url":null,"abstract":"<div><p>LiDAR-based semantic scene understanding holds a pivotal role in various applications, including remote sensing and autonomous driving. However, the majority of LiDAR segmentation models rely on extensive and densely annotated training datasets, which is extremely laborious to annotate and hinder the widespread adoption of LiDAR systems. Semi-supervised learning (SSL) offers a promising solution by leveraging only a small amount of labeled data and a larger set of unlabeled data, aiming to train robust models with desired accuracy comparable to fully supervised learning. A typical pipeline of SSL involves the initial use of labeled data to train segmentation models, followed by the utilization of predictions generated from unlabeled data, which are used as pseudo-ground truths for model retraining. However, the scarcity of labeled data limits the capture of comprehensive representations, leading to the constraints of these pseudo-ground truths in reliability. We observed that objects captured by LiDAR sensors from varying perspectives showcase diverse data characteristics due to occlusions and distance variation, and LiDAR segmentation models trained with limited labels prove susceptible to these viewpoint disparities, resulting in inaccurately predicted pseudo-ground truths across viewpoints and the accumulation of retraining errors. To address this problem, we introduce the Temporal-Selective Guided Learning (TSG-Seg) framework. TSG-Seg explores temporal cues inherent in LiDAR frames to bridge the cross-viewpoint representations, fostering consistent and robust segmentation predictions across differing viewpoints. Specifically, we first establish point-wise correspondences across LiDAR frames with different time stamps through point registration. Subsequently, reliable point predictions are selected and propagated to points from adjacent views to the current view, serving as strong and refined supervision signals for subsequent model re-training to achieve better segmentation. We conducted extensive experiments on various SSL labeling setups across multiple public datasets, including SemanticKITTI and SemanticPOSS, to evaluate the effectiveness of TSG-Seg. Our results demonstrate its competitive performance and robustness in diverse scenarios, from data-limited to data-abundant settings. Notably, TSG-Seg achieves a mIoU of 48.6% using only 5% of and 62.3% with 40% of labeled data in the sequential split on SemanticKITTI. This consistently outperforms state-of-the-art segmentation methods, including GPC and LaserMix. These findings underscore TSG-Seg’s superior capability and potential for real-world applications. The project can be found at <span><span>https://tsgseg.github.io</span><svg><path></path></svg></span>.</p></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"216 ","pages":"Pages 217-228"},"PeriodicalIF":10.6000,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ISPRS Journal of Photogrammetry and Remote Sensing","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0924271624002879","RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GEOGRAPHY, PHYSICAL","Score":null,"Total":0}
引用次数: 0

Abstract

LiDAR-based semantic scene understanding holds a pivotal role in various applications, including remote sensing and autonomous driving. However, the majority of LiDAR segmentation models rely on extensive and densely annotated training datasets, which is extremely laborious to annotate and hinder the widespread adoption of LiDAR systems. Semi-supervised learning (SSL) offers a promising solution by leveraging only a small amount of labeled data and a larger set of unlabeled data, aiming to train robust models with desired accuracy comparable to fully supervised learning. A typical pipeline of SSL involves the initial use of labeled data to train segmentation models, followed by the utilization of predictions generated from unlabeled data, which are used as pseudo-ground truths for model retraining. However, the scarcity of labeled data limits the capture of comprehensive representations, leading to the constraints of these pseudo-ground truths in reliability. We observed that objects captured by LiDAR sensors from varying perspectives showcase diverse data characteristics due to occlusions and distance variation, and LiDAR segmentation models trained with limited labels prove susceptible to these viewpoint disparities, resulting in inaccurately predicted pseudo-ground truths across viewpoints and the accumulation of retraining errors. To address this problem, we introduce the Temporal-Selective Guided Learning (TSG-Seg) framework. TSG-Seg explores temporal cues inherent in LiDAR frames to bridge the cross-viewpoint representations, fostering consistent and robust segmentation predictions across differing viewpoints. Specifically, we first establish point-wise correspondences across LiDAR frames with different time stamps through point registration. Subsequently, reliable point predictions are selected and propagated to points from adjacent views to the current view, serving as strong and refined supervision signals for subsequent model re-training to achieve better segmentation. We conducted extensive experiments on various SSL labeling setups across multiple public datasets, including SemanticKITTI and SemanticPOSS, to evaluate the effectiveness of TSG-Seg. Our results demonstrate its competitive performance and robustness in diverse scenarios, from data-limited to data-abundant settings. Notably, TSG-Seg achieves a mIoU of 48.6% using only 5% of and 62.3% with 40% of labeled data in the sequential split on SemanticKITTI. This consistently outperforms state-of-the-art segmentation methods, including GPC and LaserMix. These findings underscore TSG-Seg’s superior capability and potential for real-world applications. The project can be found at https://tsgseg.github.io.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
TSG-Seg:用于三维激光雷达点云半监督语义分割的时间选择性指导
基于激光雷达的语义场景理解在遥感和自动驾驶等各种应用中发挥着举足轻重的作用。然而,大多数激光雷达分割模型都依赖于大量密集标注的训练数据集,标注工作极其繁重,阻碍了激光雷达系统的广泛应用。半监督学习(SSL)提供了一种很有前景的解决方案,它只利用少量标注数据和更大的非标注数据集,旨在训练出稳健的模型,其预期精度可与完全监督学习相媲美。SSL 的典型流程包括首先使用标注数据训练分割模型,然后利用未标注数据生成的预测结果,将其作为伪地面真相用于模型再训练。然而,标注数据的稀缺性限制了对全面表征的捕捉,导致这些伪地面真实的可靠性受到制约。我们观察到,由于遮挡和距离变化,激光雷达传感器从不同视角捕捉到的物体呈现出不同的数据特征,而使用有限标签训练的激光雷达分割模型很容易受到这些视角差异的影响,从而导致不同视角的伪地面真值预测不准确,并积累了再训练误差。为了解决这个问题,我们引入了时间选择性指导学习(TSG-Seg)框架。TSG-Seg 利用激光雷达帧中固有的时间线索来弥合跨视点表征,从而在不同视点之间实现一致、稳健的分割预测。具体来说,我们首先通过点注册建立不同时间戳的激光雷达帧之间的点对应关系。随后,选择可靠的点预测并传播到当前视图相邻视图的点上,作为后续模型再训练的强大而精细的监督信号,以实现更好的分割。我们在多个公共数据集(包括 SemanticKITTI 和 SemanticPOSS)的各种 SSL 标签设置上进行了广泛的实验,以评估 TSG-Seg 的有效性。实验结果表明,在从数据有限到数据丰富的各种场景中,TSG-Seg 都具有极具竞争力的性能和鲁棒性。值得注意的是,在SemanticKITTI的顺序分割中,TSG-Seg仅使用5%的标记数据就实现了48.6%的mIoU,使用40%的标记数据实现了62.3%的mIoU。这始终优于最先进的分割方法,包括GPC和LaserMix。这些发现凸显了TSG-Seg在实际应用中的卓越能力和潜力。该项目的网址是 。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
ISPRS Journal of Photogrammetry and Remote Sensing
ISPRS Journal of Photogrammetry and Remote Sensing 工程技术-成像科学与照相技术
CiteScore
21.00
自引率
6.30%
发文量
273
审稿时长
40 days
期刊介绍: The ISPRS Journal of Photogrammetry and Remote Sensing (P&RS) serves as the official journal of the International Society for Photogrammetry and Remote Sensing (ISPRS). It acts as a platform for scientists and professionals worldwide who are involved in various disciplines that utilize photogrammetry, remote sensing, spatial information systems, computer vision, and related fields. The journal aims to facilitate communication and dissemination of advancements in these disciplines, while also acting as a comprehensive source of reference and archive. P&RS endeavors to publish high-quality, peer-reviewed research papers that are preferably original and have not been published before. These papers can cover scientific/research, technological development, or application/practical aspects. Additionally, the journal welcomes papers that are based on presentations from ISPRS meetings, as long as they are considered significant contributions to the aforementioned fields. In particular, P&RS encourages the submission of papers that are of broad scientific interest, showcase innovative applications (especially in emerging fields), have an interdisciplinary focus, discuss topics that have received limited attention in P&RS or related journals, or explore new directions in scientific or professional realms. It is preferred that theoretical papers include practical applications, while papers focusing on systems and applications should include a theoretical background.
期刊最新文献
Optimizing hybrid models for canopy nitrogen mapping from Sentinel-2 in Google Earth Engine A unique dielectric constant estimation for lunar surface through PolSAR model-based decomposition Unwrap-Net: A deep neural network-based InSAR phase unwrapping method assisted by airborne LiDAR data METNet: A mesh exploring approach for segmenting 3D textured urban scenes On-orbit geometric calibration of MERSI whiskbroom scanner
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1