[1]WANG Xingtao,DAI Mengli,YU Jiakuo,et al.A review of deep learning-based point cloud denoising methods[J].CAAI Transactions on Intelligent Systems,2026,21(4):849-863.[doi:10.11992/tis.202511005]
Copy

A review of deep learning-based point cloud denoising methods

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 21 Number of periods: 2026 4 Page number: 849-863 Column: 综述 Public date: 2026-07-05
Title:
A review of deep learning-based point cloud denoising methods
Author(s):
WANG Xingtao1; 2;  DAI Mengli1;  YU Jiakuo1;  LI Wenrui1;  MAN Hengyu1;  FAN Xiaopeng1; 2
1. Faculty of Computing, Harbin Institute of Technology, Harbin 150001, China;
2. Suzhou Research Institute, Harbin Institute of Technology, Suzhou 215000, China
Keywords:
3D data; point clouds; denoising; deep learning; neural networks; feature extraction; supervised learning; graph neural networks
CLC:
TP391
DOI:
10.11992/tis.202511005
Abstract:
Point clouds, as an important carrier of 3D data, are widely applied in autonomous driving, embodied intelligence, virtual reality, and other fields. However, raw point clouds are often corrupted by sensor errors, environmental interference, and reconstruction algorithm defects, resulting in noise, geometric structure distortion, and degraded performance in downstream tasks. To address these issues, numerous deep learning-based point cloud denoising methods have been proposed. This paper systematically reviews developments in this field over the past eight years. First, a five-category taxonomy for deep learning-based point cloud denoising is established, encompassing PointNet-based backbones, convolutional neural network(CNN), Transformer networks, nontraditional networks, and unsupervised learning strategies. The core mechanisms and evolutionary trajectories of these algorithms are analyzed in depth. Second, the performance of mainstream algorithms is evaluated and compared on both synthetic and real-world datasets. Finally, current challenges, such as limited generalization ability and data scarcity, are summarized, and future research directions, including the development of all-in-one models and adaptive noise modeling, are discussed. The analysis presented in this paper provides a reference for the taxonomy, performance evaluation, and future improvement of deep learning-based point cloud denoising methods.
References:
[1] 王昶畅, 江坤, 姜凯, 等. 基于反馈的迭代采样高噪声点云去噪框架[J]. 图学学报, 2025, 46(3): 614-623 WANG Changchang, JIANG Kun, JIANG Kai, et al. Feedback-based iterative sampling denoising framework for point clouds with high-level noise[J]. Journal of graphics, 2025, 46(3): 614-623
[2] 宋小英, 徐征, 于长龙. 三维激光点云数据去噪方法研究[J]. 测绘与空间地理信息, 2025, 48(5): 202-204, 211 SONG Xiaoying, XU Zheng, YU Changlong. Research on the denoising method of 3D laser point cloud data[J]. Geomatics & spatial information technology, 2025, 48(5): 202-204, 211
[3] 曲金博, 王岩, 赵琪. DBSCAN 聚类和改进的双边滤波算法在点云去噪中的应用[J]. 测绘通报, 2019(11): 89-92 QU Jinbo, WANG Yan, ZHAO Qi. Application of DBSCAN clustering and improved bilateral filtering algorithm in point cloud denoising[J]. Bulletin of surveying and mapping, 2019(11): 89-92
[4] 王兴涛. 基于深度学习的点云去噪研究[D]. 哈尔滨: 哈尔滨工业大学, 2022. WANG Xingtao. Research on deep-learning point cloud denoising[D]. Harbin: Harbin Institute of Technology, 2022.
[5] ZHOU Lang, SUN Guoxing, LI Yong, et al. Point cloud denoising review: from classical to deep learning-based approaches[J]. Graphical models, 2022, 121: 101140
[6] 肖俊, 石光田. 三维点云去噪技术[J]. 中国科学院大学学报, 2023, 40(5): 577-595 XIAO Jun, SHI Guangtian. Three-dimensional point cloud denoising[J]. Journal of University of Chinese Academy of Sciences, 2023, 40(5): 577-595
[7] 吴一全, 陈慧娴, 张耀. 基于深度学习的三维点云处理方法研究进展[J]. 中国激光, 2024, 51(5): 143-165 WU Yiquan, CHEN Huixian, ZHANG Yao. Review of 3D point cloud processing methods based on deep learning[J]. Chinese journal of lasers, 2024, 51(5): 143-165
[8] LI Ying, SHENG Huankun. A single-stage point cloud cleaning network for outlier removal and denoising[J]. Pattern recognition, 2023, 138: 109366
[9] WANG Ziwei, SUN Wei, TIAN Linyang. 3D point cloud denoising based on hybrid attention mechanism and score matching[C]//Proceedings of the 2022 5th International Conference on Artificial Intelligence and Pattern Recognition. Xiamen: ACM, 2022.
[10] WANG Junbo, LI Ying. Transformer-based point cloud denoising network[C]//2023 4th International Conference on Big Data & Artificial Intelligence & Software Engineering. Nanjing: IEEE, 2023.
[11] REN Zhiyuan, KIM M, LIU Feng, et al. TIGER: time-varying denoising model for 3D point cloud generation with diffusion process[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2024.
[12] 孔大力, 江平. 基于Transformer和多尺度的点云去噪[J]. 大学数学, 2023, 39(4): 7-15 KONG Dali, JIANG Ping. Point cloud denoising based on Transformer and multi scale[J]. College mathematics, 2023, 39(4): 7-15
[13] CHARLES R Q, HAO Su, MO Kaichun, et al. PointNet: deep learning on point sets for 3D classification and segmentation[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017.
[14] 李美佳, 于泽宽, 刘晓, 等. 点云算法在医学领域的研究进展[J]. 中国图象图形学报, 2020, 25(10): 2013-2023 LI Meijia, YU Zekuan, LIU Xiao, et al. Progress of point cloud algorithm in medical field[J]. Journal of image and graphics, 2020, 25(10): 2013-2023
[15] QI C R, YI Li, SU Hao, et al. PointNet++: deep hierarchical feature learning on point sets in a metric space[C]//Proceedings of the 30th International Conference on Neural Information Processing Systems. Long Beach: Curran Associates, 2017.
[16] GUERRERO P, KLEIMAN Y, OVSJANIKOV M, et al. PCPNet learning local shape properties from raw point clouds[J]. Computer graphics forum, 2018, 37(2): 75-85
[17] YU Lequan, LI Xianzhi, FU C W, et al. EC-Net: an edge-aware point set consolidation network[C]//Computer Vision. Cham: Springer, 2018.
[18] DUAN Chaojing, CHEN Siheng, KOVACEVIC J. 3D point cloud denoising via deep neural network based local surface estimation[C]//2019 IEEE International Conference on Acoustics, Speech and Signal Processing. Brighton: IEEE, 2019.
[19] RAKOTOSAONA M J, LA BARBERA V, GUERRERO P, et al. PointCleanNet: learning to denoise and remove outliers from dense point clouds[J]. Computer graphics forum, 2020, 39(1): 185-203
[20] ZHANG Dongbo, LU Xuequan, QIN Hong, et al. Pointfilter: point cloud filtering via encoder-decoder modeling[J]. IEEE transactions on visualization and computer graphics, 2021, 27(3): 2015-2027
[21] HUANG Anyi, XIE Qian, WANG Zhoutao, et al. MODNet: multi-offset point cloud denoising network customized for multi-scale patches[J]. Computer graphics forum, 2022, 41(7): 109-119
[22] WANG Xingtao, FAN Xiaopeng, ZHAO Debin. PointFilterNet: a filtering network for point cloud denoising[J]. IEEE transactions on circuits and systems for video technology, 2023, 33(3): 1276-1290
[23] WANG Xingtao, CUI Wenxue, XIONG Ruiqin, et al. FCNet: learning noise-free features for point cloud denoising[J]. IEEE transactions on circuits and systems for video technology, 2023, 33(11): 6288-6301
[24] WEI Mingqiang, WEI Zeyong, ZHOU Haoran, et al. AGConv: adaptive graph convolution on 3D point clouds[J]. IEEE transactions on pattern analysis and machine intelligence, 2023, 45(8): 9374-9392
[25] ROVERI R, ?ZTIRELI A C, PANDELE I, et al. PointProNets: consolidation of point clouds with convolutional neural networks[J]. Computer graphics forum, 2018, 37(2): 87-99
[26] PISTILLI F, FRACASTORO G, VALSESIA D, et al. Learning graph-convolutional representations for point cloud denoising[C]//Computer Vision. Cham: Springer, 2020.
[27] PISTILLI F, FRACASTORO G, VALSESIA D, et al. Learning robust graph-convolutional representations for point cloud denoising[J]. IEEE journal of selected topics in signal processing, 2021, 15(2): 402-414
[28] CHEN Zhaowei, LI Peng, WEI Zeyong, et al. GeoGCN: geometric dual-domain graph convolution network for point cloud denoising[C]//2023 IEEE International Conference on Acoustics, Speech and Signal Processing. Rhodes Island: IEEE, 2023.
[29] DE SILVA EDIRIMUNI D, LU Xuequan, SHAO Zhiwen, et al. IterativePFN: true iterative point cloud filtering[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Vancouver: IEEE, 2023.
[30] YI Cheng, WEI Zeyong, QIU Jingbo, et al. PN-Internet: point-and-normal interactive network for noisy point clouds[J]. IEEE transactions on geoscience and remote sensing, 2024, 62: 1-11
[31] LU Dening, LU Xuequan, SUN Yangxing, et al. Deep feature-preserving normal estimation for point cloud filtering[J]. Computer-aided design, 2020, 125: 102860
[32] WEI Mingqiang, CHEN Honghua, ZHANG Yingkui, et al. GeoDualCNN: geometry-supporting dual convolutional neural network for noisy point clouds[J]. IEEE transactions on visualization and computer graphics, 2023, 29(2): 1357-1370
[33] 黄东福, 刘立群. 基于图神经网络的异源图像配准方法综述[J]. 软件工程, 2025, 28(1): 1-7 HUANG Dongfu, LIU Liqun. Overview of heterogeneous image registration methods based on graph neural networks[J]. Software engineering, 2025, 28(1): 1-7
[34] 杨珊珊. 基于特征融合的点云配准技术研究[D]. 郑州: 郑州大学, 2022. YANG Shanshan. Research of point cloud registration technologybased on feature fusion[D]. Zhengzhou: Zhengzhou University, 2022.
[35] WANG Yue, SUN Yongbin, LIU Ziwei, et al. Dynamic graph CNN for learning on point clouds[J]. ACM transactions on graphics, 2019, 38(5): 1-12
[36] LUO Shitong, HU Wei. Differentiable manifold reconstruction for point cloud denoising[C]//Proceedings of the 28th ACM International Conference on Multimedia. Seattle: ACM, 2020.
[37] LUO Shitong, HU Wei. Score-based point cloud denoising[C]//2021 IEEE/CVF International Conference on Computer Vision. Montreal: IEEE, 2021.
[38] LI Zhenglei, PAN Weigang, WANG Shuxin, et al. A point cloud denoising network based on manifold in an unknown noisy environment[J]. Infrared physics & technology, 2023, 132: 104735
[39] WANG Weijia, PAN Wei, LIU Xiao, et al. Random screening-based feature aggregation for point cloud denoising[J]. Computers & graphics, 2023, 116: 64-72
[40] YANG Wenming, HE Zhouyan, SONG Yang, et al. 3D point cloud denoising method based on global feature guidance[J]. The visual computer, 2024, 40(9): 6137-6153
[41] MAO Aihua, YAN Biao, MA Zijing, et al. Denoising point clouds in latent space via graph convolution and invertible neural network[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2024.
[42] 孔大力. 基于多尺度Transformer网络的点云去噪研究[D]. 合肥: 合肥工业大学, 2023. KONG Dali. The research of point cloud denoising based on multi-scale transformer network[D]. Hefei: Hefei University of Technology, 2023.
[43] 李玉洁, 马子航, 王艺甫, 等. 视觉Transformer (ViT) 发展综述[J]. 计算机科学, 2025, 52(1): 194–209. LI Yujie, MA Zihang, WANG Yifu, et al. Survey of vision transformers(ViT)[J]. 2025, 52(1): 194–209.
[44] XU Xueli, GENG Guohua, CAO Xin, et al. TDNet: transformer-based network for point cloud denoising[J]. Applied optics, 2022, 61(6): C80
[45] ZHU Xusheng, MA Shuai, CHEN Daixin, et al. MSaD-Net: a mix self-attention networks for 3D point cloud denoising[J]. IEEE photonics journal, 2023, 15(3): 1-7
[46] MAO Aihua, DU Zihui, WEN Yuhui, et al. PD-flow: a point cloud denoising framework withNormalizing flows[C]//Computer Vision – ECCV 2022. Cham: Springer, 2022.
[47] CHEN Honghua, WEI Zeyong, LI Xianzhi, et al. RePCD-Net: feature-aware recurrent point cloud denoising network[J]. International journal of computer vision, 2022, 130(3): 615-629
[48] SHENG Huankun, LI Ying. Denoising point clouds with fewer learnable parameters[J]. Computer-aided design, 2024, 172: 103708
[49] 许同骏, 许杰. 基于深度强化学习的推荐算法的构建研究[J]. 电脑知识与技术, 2025, 21(9): 33-37 XU Tongjun, XU Jie. Research on the construction of recommendation algorithms based on deep reinforcement learning[J]. Computer knowledge and technology, 2025, 21(9): 33-37
[50] WEI Zeyong, CHEN Honghua, NAN Liangliang, et al. PathNet: path-selective point cloud denoising[J]. IEEE transactions on pattern analysis and machine intelligence, 2024, 46(6): 4426-4442
[51] VOGEL M, TATENO K, POLLEFEYS M, et al. P2P-bridge: diffusion bridges for3D point cloud denoising[C]//Computer Vision – ECCV 2024. Cham: Springer, 2025.
[52] LIU Zheng, ZHOU Weijie, GUO Chuchen, et al. PyramidPCD: a novel pyramid network for point cloud denoising[J]. Pattern recognition, 2025, 161: 111228
[53] DU Qidong. 3D point cloud registration denoising method for human motion image using deep learning algorithm[J]. Multimedia systems, 2020, 26(1): 75-82
[54] WU Jianeng, XIANG Lirong, YOU Hui, et al. Plant-Denoising-Net (PDN): a plant point cloud denoising network based on density gradient field learning[J]. ISPRS journal of photogrammetry and remote sensing, 2024, 210: 282-299
[55] LIN Runheng, HU Hao, WEN Zhikun, et al. Research on denoising and segmentation algorithm application of pigs’ point cloud based on DBSCAN and PointNet[C]//2021 IEEE International Workshop on Metrology for Agriculture and Forestry. Trento-Bolzano: IEEE, 2021.
[56] GAN Xiaohang, TAN Libin, WANG Xiaoyi. Application study of PointNet++ and hybrid filtering based point cloud denoising and segmentation algorithms for Blisk[C]//2022 5th World Conference on Mechanical Engineering and Intelligent Manufacturing. Ma’anshan: IEEE, 2022.
[57] ZUO L L, ZHANG J, DING S L, et al. Attention mechanism-based deep learning denoising of scanned point cloud for rocket tank panel[C]//2023 IEEE International Conference on Industrial Engineering and Engineering Management. Singapore: IEEE, 2023.
[58] SEPP?NEN A, OJALA R, TAMMI K. Self-supervised multi-echo point cloud denoising in snowfall[J]. Pattern recognition letters, 2024, 185: 52-58
[59] LIU Junyu, REN Jianfeng, SUN Hongliang, et al. Face recognition on point cloud with CGAN-Top for denoising[C]//2023 IEEE International Conference on Acoustics, Speech and Signal Processing. Rhodes Island: IEEE, 2023.
[60] 侯广哲, 秦贵和, 梁艳花. 基于下采样的自监督点云去噪方法[J]. 吉林大学学报(理学版), 2024, 62(1): 100-105 HOU Guangzhe, QIN Guihe, LIANG Yanhua. Self-supervised point cloud denoising method based on downsampling[J]. Journal of Jilin University (science edition), 2024, 62(1): 100-105
[61] LEHTINEN J, MUNKBERG J, HASSELGREN J, et al. Noise2Noise: learning image restoration without clean data[C]//Proceedings of the 35th International Conference on Machine Learning. Brookline: PMLR, 2018.
[62] HERMOSILLA P, RITSCHEL T, ROPINSKI T. Total denoising: unsupervised learning of 3D point cloud cleaning[C]//2019 IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019.
[63] CHEN Siheng, DUAN Chaojing, YANG Yaoqing, et al. Deep unsupervised learning of 3D point clouds via graph topology inference and filtering[J]. IEEE transactions on image processing, 2020, 29: 3183-3198
[64] WANG Weijia, LIU Xiao, ZHOU Hailing, et al. Noise4Denoise: leveraging noise for unsupervised point cloud denoising[J]. Computational visual media, 2024, 10(4): 659-669
[65] MA Baorui, LIU Yushen, HAN Zhizhong. Learning signed distance functions from noisy 3D point clouds via noise to noise mapping[C]//Proceedings of the 40th International Conference on Machine Learning. Brookline: PMLR, 2023.
[66] ZHOU Junsheng, MA Baorui, LIU Yushen, et al. Fast learning of signed distance functions from noisy point clouds via noise to noise mapping[J]. IEEE transactions on pattern analysis and machine intelligence, 2024, 46(12): 8936-8953
[67] SU Zhiyong, WANG Changchang, JIANG Kun, et al. SITF: a self-supervised iterative training framework for point cloud denoising[J]. Computer-aided design, 2025, 179: 103812
[68] DU Yi, ZHAO Zhipeng, SU Shaoshu, et al. SuperPC: a single diffusion model for point cloud completion, upsampling, denoising, and colorization[C]//2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville: IEEE, 2025.
Similar References:

Memo

-

Last Update: 1900-01-01

Copyright © CAAI Transactions on Intelligent Systems