DC3DCD: Unsupervised learning for multiclass 3D point cloud change detection

计算机科学 变更检测 人工智能 点云 无监督学习 分割 背景(考古学) 深度学习 机器学习 任务(项目管理) 模式识别(心理学) 监督学习 人工神经网络 古生物学 管理 经济 生物
作者
Iris de Gélis,Sébastien Lefèvre,Thomas Corpetti
出处
期刊:Isprs Journal of Photogrammetry and Remote Sensing 卷期号:206: 168-183
标识
DOI:10.1016/j.isprsjprs.2023.10.022
摘要

In a constant evolving world, change detection is of prime importance to keep updated maps. To better sense areas with complex geometry (urban areas in particular), considering 3D data appears to be an interesting alternative to classical 2D images. In this context, 3D point clouds (PCs), whether obtained through LiDAR or photogrammetric techniques, provide valuable information. While recent studies showed the considerable benefit of using deep learning-based methods to detect and characterize changes into raw 3D PCs, these studies rely on large annotated training data to obtain accurate results. The collection of these annotations are tricky and time-consuming. The availability of unsupervised or weakly supervised approaches is then of prime interest. In this paper, we propose an unsupervised method, called DeepCluster 3D Change Detection (DC3DCD), to detect and categorize multiclass changes at point level. We classify our approach in the unsupervised family given the fact that we extract in a completely unsupervised way a number of clusters associated with potential changes. Let us precise that in the end of the process, the user has only to assign a label to each of these clusters to derive the final change map. Our method builds upon the DeepCluster approach, originally designed for image classification, to handle complex raw 3D PCs and perform change segmentation task. An assessment of the method on both simulated and real public dataset is provided. The proposed method allows to outperform fully-supervised traditional machine learning algorithm and to be competitive with fully-supervised deep learning networks applied on rasterization of 3D PCs with a mean of IoU over classes of change of 57.06% and 66.69% for the simulated and the real datasets, respectively. The code is available at https://github.com/idegelis/torch-points3d-dc3dcd.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
aryecatcher发布了新的文献求助10
2秒前
共享精神应助peach采纳,获得10
2秒前
大黄鸭的小黄人完成签到,获得积分10
2秒前
pengzzZZ发布了新的文献求助10
3秒前
阿胡发布了新的文献求助10
3秒前
Orange应助科研通管家采纳,获得10
4秒前
上官若男应助科研通管家采纳,获得10
4秒前
CHENG_2025应助科研通管家采纳,获得10
4秒前
ding应助科研通管家采纳,获得10
5秒前
5秒前
斯文败类应助科研通管家采纳,获得10
5秒前
沐青应助科研通管家采纳,获得30
5秒前
传奇3应助科研通管家采纳,获得10
5秒前
科研通AI2S应助科研通管家采纳,获得30
5秒前
乐乐应助科研通管家采纳,获得10
5秒前
乐观小之应助科研通管家采纳,获得10
5秒前
情怀应助科研通管家采纳,获得10
5秒前
NexusExplorer应助科研通管家采纳,获得10
5秒前
ding应助科研通管家采纳,获得10
5秒前
6秒前
宇宙第一帅发布了新的文献求助200
7秒前
7秒前
7秒前
zhou默完成签到,获得积分10
7秒前
FCL完成签到,获得积分10
9秒前
10秒前
10秒前
景行行止发布了新的文献求助10
11秒前
11秒前
高兴的羊完成签到,获得积分10
12秒前
星辰大海应助跳跳妈妈采纳,获得30
13秒前
共享精神应助秋不落棠采纳,获得10
14秒前
1234发布了新的文献求助10
14秒前
peach发布了新的文献求助10
14秒前
圆圆发布了新的文献求助20
15秒前
冷傲的电源完成签到,获得积分10
15秒前
leslie发布了新的文献求助10
16秒前
英姑应助like1994采纳,获得10
16秒前
16秒前
杨欢欢发布了新的文献求助10
16秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Cognitive Neuroscience: The Biology of the Mind 1000
Technical Brochure TB 814: LPIT applications in HV gas insulated switchgear 1000
Toward a Combinatorial Approach for the Prediction of IgG Half-Life and Clearance 500
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Picture Books with Same-sex Parented Families: Unintentional Censorship 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3969940
求助须知:如何正确求助?哪些是违规求助? 3514642
关于积分的说明 11175298
捐赠科研通 3249947
什么是DOI,文献DOI怎么找? 1795178
邀请新用户注册赠送积分活动 875617
科研通“疑难数据库(出版商)”最低求助积分说明 804891