{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/volumetric-and-multi-view-cnns-for-object","title":"Volumetric and Multi-View CNNs for Object Classification on 3D Data","arxiv_id":"1604.03265","date":"2016-04-12","proceeding":"CVPR 2016 6","authors":["Charles R. Qi","Hao Su","Matthias Niessner","Angela Dai","Mengyuan Yan","Leonidas J. Guibas"],"abstract":"3D shape models are becoming widely available and easier to capture, making\navailable 3D information crucial for progress in object classification. Current\nstate-of-the-art methods rely on CNNs to address this problem. Recently, we\nwitness two types of CNNs being developed: CNNs based upon volumetric\nrepresentations versus CNNs based upon multi-view representations. Empirical\nresults from these two types of CNNs exhibit a large gap, indicating that\nexisting volumetric CNN architectures and approaches are unable to fully\nexploit the power of 3D representations. In this paper, we aim to improve both\nvolumetric CNNs and multi-view CNNs according to extensive analysis of existing\napproaches. To this end, we introduce two distinct network architectures of\nvolumetric CNNs. In addition, we examine multi-view CNNs, where we introduce\nmulti-resolution filtering in 3D. Overall, we are able to outperform current\nstate-of-the-art methods for both volumetric CNNs and multi-view CNNs. We\nprovide extensive experiments designed to evaluate underlying design choices,\nthus providing a better understanding of the space of methods available for\nobject classification on 3D data.","url_abs":"http://arxiv.org/abs/1604.03265v2","url_pdf":"http://arxiv.org/pdf/1604.03265v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"volumetric-and-multi-view-cnns-for-object","repo_url":"https://github.com/LONG-9621/3DCNN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"volumetric-and-multi-view-cnns-for-object","repo_url":"https://github.com/charlesq34/3dcnn.torch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"3d-object-recognition","task_name":"3D Object Recognition"},{"task_slug":"3d-point-cloud-classification","task_name":"3D Point Cloud Classification"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-object-recognition-on-modelnet40","task":"3D Object Recognition","dataset":"ModelNet40","model":"MVCNN-MultiRes","rank_in_archive_order":3,"of":6,"metrics":{"Accuracy":"93.8%"},"uses_additional_data":false},{"leaderboard":"/sota/3d-point-cloud-classification-on-modelnet40","task":"3D Point Cloud Classification","dataset":"ModelNet40","model":"Subvolume","rank_in_archive_order":106,"of":111,"metrics":{"Overall Accuracy":"89.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1604.03265","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}