{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T15:25:12Z","timestamp":1759332312873,"version":"3.41.2"},"reference-count":41,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,1,6]],"date-time":"2023-01-06T00:00:00Z","timestamp":1672963200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100006769","name":"Russian Science Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100006769","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Robot. AI"],"abstract":"<jats:p>There are a large number of publicly available datasets of 3D data, they generally suffer from some drawbacks, such as small number of data samples, and class imbalance. Data augmentation is a set of techniques that aim to increase the size of datasets and solve such defects, and hence to overcome the problem of overfitting when training a classifier. In this paper, we propose a method to create new synthesized data by converting complete meshes into occluded 3D point clouds similar to those in real-world datasets. The proposed method involves two main steps, the first one is hidden surface removal (HSR), where the occluded parts of objects surfaces from the viewpoint of a camera are deleted. A low-complexity method has been proposed to implement HSR based on occupancy grids. The second step is a random sampling of the detected visible surfaces. The proposed two-step method is applied to a subset of ModelNet40 dataset to create a new dataset, which is then used to train and test three different deep-learning classifiers (VoxNet, PointNet, and 3DmFV). We studied classifiers performance as a function of the camera elevation angle. We also conducted another experiment to show how the newly generated data samples can improve the classification performance when they are combined with the original data during training process. Simulation results show that the proposed method enables us to create a large number of new data samples with a small size needed for storage. Results also show that the performance of classifiers is highly dependent on the elevation angle of the camera. In addition, there may exist some angles where performance degrades significantly. Furthermore, data augmentation using our created data improves the performance of classifiers not only when they are tested on the original data, but also on real data.<\/jats:p>","DOI":"10.3389\/frobt.2022.1077895","type":"journal-article","created":{"date-parts":[[2023,1,6]],"date-time":"2023-01-06T17:48:40Z","timestamp":1673027320000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["A method to create real-like point clouds for 3D object classification"],"prefix":"10.3389","volume":"9","author":[{"given":"Vladimir Ivanovich","family":"Syryamkin","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Majdi","family":"Msallam","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Semen Aleksandrovich","family":"Klestov","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2023,1,6]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"3145","DOI":"10.1109\/lra.2018.2850061","article-title":"3DmFV: Three-dimensional point cloud classification in real-time using convolutional neural networks","volume":"3","author":"Ben-Shabat","year":"2018","journal-title":"IEEE Robot. Autom. Lett."},{"key":"B2","first-page":"6302","article-title":"POCO: Point convolution for surface reconstruction","author":"Boulch","year":"2022"},{"key":"B3","unstructured":"ShapeNet: An information-rich 3D model repository\n            ChangA. X.\n            FunkhouserT.\n            GuibasL.\n            HanrahanP.\n            HuangQ.\n            LiZ.\n            YuF.\n          2015"},{"key":"B4","first-page":"2514","article-title":"Deep global registration","author":"Choy","year":"2020"},{"key":"B5","doi-asserted-by":"publisher","first-page":"412","DOI":"10.1109\/tvcg.2003.1207447","article-title":"A survey of visibility for walkthrough applications","volume":"9","author":"Cohen-Or","year":"2003","journal-title":"IEEE Trans. Vis. Comput. Graph."},{"key":"B6","first-page":"1","article-title":"Unsupervised feature learning for classification of outdoor 3d scans","volume-title":"Australasian conference on robotics and automation","author":"De Deuge","year":"2013"},{"key":"B7","doi-asserted-by":"publisher","first-page":"7503","DOI":"10.1039\/c9sc02097g","article-title":"A generalized deep learning approach for local structure identification in molecular simulations","volume":"10","author":"DeFever","year":"2019","journal-title":"Chem. Sci."},{"key":"B8","doi-asserted-by":"publisher","first-page":"237","DOI":"10.1109\/34.982903","article-title":"Vision for mobile robot navigation: A survey","volume":"24","author":"DeSouza","year":"2002","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B9","doi-asserted-by":"publisher","first-page":"4338","DOI":"10.1109\/tpami.2020.3005434","article-title":"Deep learning for 3d point clouds: A survey","volume":"43","author":"Guo","year":"2020","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B10","first-page":"1281","article-title":"Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms","author":"Heyburn","year":"2018"},{"volume-title":"Computer graphics: Principles and practice","year":"2014","author":"Hughes","key":"B11"},{"key":"B12","unstructured":"Synthetic Data -- what, why and how?\n            JordonJ.\n            SzpruchL.\n            HoussiauF.\n            BottarelliM.\n            CherubinG.\n            MapleC.\n          2022"},{"key":"B13","unstructured":"Physics-informed PointNet: A deep learning solver for steady-state incompressible flows and thermal fields on multiple sets of irregular geometries\n            KashefiA.\n            MukerjiT.\n          2022"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1063\/5.0063904","article-title":"Point-cloud deep learning of porous media for permeability prediction","volume":"33","author":"Kashefi","year":"2021","journal-title":"Phys. Fluids"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1063\/5.0033376","article-title":"A point-cloud deep learning framework for prediction of fluid flow fields on irregular geometries","volume":"33","author":"Kashefi","year":"2021","journal-title":"Phys. Fluids"},{"key":"B16","doi-asserted-by":"publisher","first-page":"466","DOI":"10.3390\/sym14030466","article-title":"Mesh clustering and reordering based on normal locality for efficient rendering","volume":"14","author":"Kim","year":"2022","journal-title":"Symmetry"},{"key":"B17","doi-asserted-by":"publisher","first-page":"792","DOI":"10.1109\/70.88100","article-title":"Stereo vision and navigation in buildings for mobile robots","volume":"5","author":"Kriegman","year":"1989","journal-title":"IEEE Trans. Robot. Autom."},{"key":"B18","doi-asserted-by":"publisher","first-page":"4188","DOI":"10.3390\/s19194188","article-title":"Deep learning on point clouds and its application: A survey","volume":"19","author":"Liu","year":"2019","journal-title":"Sensors"},{"key":"B19","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1023\/b:auro.0000047286.62481.1d","article-title":"Obstacle detection and terrain classification for autonomous off-road navigation","volume":"18","author":"Manduchi","year":"2005","journal-title":"Aut. robots"},{"key":"B20","first-page":"637","article-title":"Real time obstacle detection for mobile robot navigation using stereo vision","author":"Mane","year":"2016"},{"key":"B21","first-page":"922","article-title":"Voxnet: A 3d convolutional neural network for real-time object recognition","author":"Maturana","year":"2015"},{"key":"B22","first-page":"3103","article-title":"PointView-GCN: 3D shape classification with multi-view point clouds","author":"Mohammadi","year":"2021"},{"key":"B23","article-title":"Dec RL-PointClouds \u2013 A new synthetic dataset to classify 3D objects","volume-title":"IV ?\u0435\u0436\u0434\u0443\u043d\u0430\u0440\u043e\u0434\u043d\u044b\u0439 \u0444\u043e\u0440\u0443\u043c \u00ab?\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0443\u0430\u043b\u044c\u043d\u044b\u0435 \u0441\u0438\u0441\u0442\u0435\u043c\u044b 4-\u0439 \u043f\u0440\u043e\u043c\u044b\u0448\u043b\u0435\u043d\u043d\u043e\u0439","author":"Msallam","year":""},{"key":"B24","doi-asserted-by":"publisher","first-page":"012013","DOI":"10.1088\/1742-6596\/1843\/1\/012013","article-title":"Improving a device for identifying and marking parts of territory with chemical and radioactive contamination journal of physics: Conferen","volume":"1843","author":"Msallam","year":"2021","journal-title":"J. Phys. Conf. Ser."},{"key":"B25","first-page":"260","article-title":"Point clouds - a method to represent objects in 3D space","author":"Msallam","year":""},{"key":"B26","first-page":"652","article-title":"Pointnet: Deep learning on point sets for 3d classification and segmentation","author":"Qi","year":""},{"key":"B27","article-title":"PointNet++: Deep hierarchical feature learning on point sets in a metric space","volume":"30","author":"Qi","year":"","journal-title":"Adv. neural Inf. Process. Syst."},{"key":"B28","first-page":"95","article-title":"Coal mine robot binocular vision recognition system based on fuzzy neural network","author":"Shang","year":"2015"},{"key":"B29","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1186\/s40537-019-0197-0","article-title":"A survey on image data augmentation for deep learning","volume":"6","author":"Shorten","year":"2019","journal-title":"J. Big Data"},{"key":"B30","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1007\/s00138-007-0097-8","article-title":"Retrieving articulated 3-D models using medial surfaces","volume":"19","author":"Siddiqi","year":"2008","journal-title":"Mach. Vis. Appl."},{"volume-title":"Precalculus: Mathematics for calculus","year":"2015","author":"Stewart","key":"B31"},{"key":"B32","unstructured":"Benchmarking robustness of 3d point cloud recognition against common corruptions\n            SunJ.\n            ZhangQ.\n            KailkhuraB.\n            YuZ.\n            XiaoC.\n            MaoZ. M.\n          2022"},{"volume-title":"Digital X-ray tomography","year":"2020","author":"Syryamkin","key":"B33"},{"key":"B34","doi-asserted-by":"crossref","first-page":"538","DOI":"10.5772\/6155","article-title":"Image processing for next-generation robots","volume-title":"Computer vision","author":"Sziebig","year":"2008"},{"key":"B35","first-page":"6411","article-title":"KPConv: Flexible and deformable convolution for point clouds","author":"Thomas","year":"2019"},{"key":"B36","first-page":"1912","article-title":"3D ShapeNets: A deep representation for volumetric shapes","author":"Wu","year":"2015"},{"key":"B37","first-page":"5589","article-title":"PointASNL: Robust point clouds processing using nonlocal neural networks with adaptive sampling","author":"Yan","year":"2020"},{"key":"B38","unstructured":"Boosting 3D object detection via object-focused image fusion\n            YangH.\n            ShiC.\n            ChenY.\n            WangL.\n          2022"},{"key":"B39","doi-asserted-by":"publisher","first-page":"130561","DOI":"10.1109\/access.2020.3009387","article-title":"Coal mine rescue robots based on binocular vision: A review of the state of the art","volume":"8","author":"Zhai","year":"2020","journal-title":"IEEE Access"},{"key":"B40","unstructured":"On automatic data augmentation for 3D point cloud classification\n            ZhangW.\n            XuX.\n            LiuF.\n            FooC. S.\n          2021"},{"key":"B41","first-page":"506","article-title":"Design of vision-based obstacle crossing of high-voltage line inspection robot","author":"Zhu","year":"2016"}],"container-title":["Frontiers in Robotics and AI"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2022.1077895\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,6]],"date-time":"2023-01-06T17:48:50Z","timestamp":1673027330000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2022.1077895\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,1,6]]},"references-count":41,"alternative-id":["10.3389\/frobt.2022.1077895"],"URL":"https:\/\/doi.org\/10.3389\/frobt.2022.1077895","relation":{},"ISSN":["2296-9144"],"issn-type":[{"type":"electronic","value":"2296-9144"}],"subject":[],"published":{"date-parts":[[2023,1,6]]},"article-number":"1077895"}}