{"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/multi-view-convolutional-neural-networks-for","title":"Multi-view Convolutional Neural Networks for 3D Shape Recognition","arxiv_id":"1505.00880","date":"2015-05-05","proceeding":"ICCV 2015 12","authors":["Hang Su","Subhransu Maji","Evangelos Kalogerakis","Erik Learned-Miller"],"abstract":"A longstanding question in computer vision concerns the representation of 3D\nshapes for recognition: should 3D shapes be represented with descriptors\noperating on their native 3D formats, such as voxel grid or polygon mesh, or\ncan they be effectively represented with view-based descriptors? We address\nthis question in the context of learning to recognize 3D shapes from a\ncollection of their rendered views on 2D images. We first present a standard\nCNN architecture trained to recognize the shapes' rendered views independently\nof each other, and show that a 3D shape can be recognized even from a single\nview at an accuracy far higher than using state-of-the-art 3D shape\ndescriptors. Recognition rates further increase when multiple views of the\nshapes are provided. In addition, we present a novel CNN architecture that\ncombines information from multiple views of a 3D shape into a single and\ncompact shape descriptor offering even better recognition performance. The same\narchitecture can be applied to accurately recognize human hand-drawn sketches\nof shapes. We conclude that a collection of 2D views can be highly informative\nfor 3D shape recognition and is amenable to emerging CNN architectures and\ntheir derivatives.","url_abs":"http://arxiv.org/abs/1505.00880v3","url_pdf":"http://arxiv.org/pdf/1505.00880v3.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":[],"tasks":[{"task_slug":"3d-point-cloud-classification","task_name":"3D Point Cloud Classification"},{"task_slug":"3d-shape-recognition","task_name":"3D Shape Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-point-cloud-classification-on-modelnet40","task":"3D Point Cloud Classification","dataset":"ModelNet40","model":"MVCNN","rank_in_archive_order":104,"of":111,"metrics":{"Overall Accuracy":"90.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1505.00880","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}