{"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/surfconv-bridging-3d-and-2d-convolution-for","title":"SurfConv: Bridging 3D and 2D Convolution for RGBD Images","arxiv_id":"1812.01519","date":"2018-12-04","proceeding":"CVPR 2018 6","authors":["Hang Chu","Wei-Chiu Ma","Kaustav Kundu","Raquel Urtasun","Sanja Fidler"],"abstract":"We tackle the problem of using 3D information in convolutional neural\nnetworks for down-stream recognition tasks. Using depth as an additional\nchannel alongside the RGB input has the scale variance problem present in image\nconvolution based approaches. On the other hand, 3D convolution wastes a large\namount of memory on mostly unoccupied 3D space, which consists of only the\nsurface visible to the sensor. Instead, we propose SurfConv, which \"slides\"\ncompact 2D filters along the visible 3D surface. SurfConv is formulated as a\nsimple depth-aware multi-scale 2D convolution, through a new Data-Driven Depth\nDiscretization (D4) scheme. We demonstrate the effectiveness of our method on\nindoor and outdoor 3D semantic segmentation datasets. Our method achieves\nstate-of-the-art performance with less than 30% parameters used by the 3D\nconvolution-based approaches.","url_abs":"http://arxiv.org/abs/1812.01519v1","url_pdf":"http://arxiv.org/pdf/1812.01519v1.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":"surfconv-bridging-3d-and-2d-convolution-for","repo_url":"https://github.com/chuhang/SurfConv","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"3d-semantic-segmentation","task_name":"3D Semantic Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"3d-convolution","method_name":"3D Convolution"},{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.01519","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}