{"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/rgbd-based-dimensional-decomposition-residual","title":"RGBD Based Dimensional Decomposition Residual Network for 3D Semantic Scene Completion","arxiv_id":"1903.00620","date":"2019-03-02","proceeding":"CVPR 2019 6","authors":["Jie Li","Yu Liu","Dong Gong","Qinfeng Shi","Xia Yuan","Chunxia Zhao","Ian Reid"],"abstract":"RGB images differentiate from depth images as they carry more details about\nthe color and texture information, which can be utilized as a vital\ncomplementary to depth for boosting the performance of 3D semantic scene\ncompletion (SSC). SSC is composed of 3D shape completion (SC) and semantic\nscene labeling while most of the existing methods use depth as the sole input\nwhich causes the performance bottleneck. Moreover, the state-of-the-art methods\nemploy 3D CNNs which have cumbersome networks and tremendous parameters. We\nintroduce a light-weight Dimensional Decomposition Residual network (DDR) for\n3D dense prediction tasks. The novel factorized convolution layer is effective\nfor reducing the network parameters, and the proposed multi-scale fusion\nmechanism for depth and color image can improve the completion and segmentation\naccuracy simultaneously. Our method demonstrates excellent performance on two\npublic datasets. Compared with the latest method SSCNet, we achieve 5.9% gains\nin SC-IoU and 5.7% gains in SSC-IOU, albeit with only 21% network parameters\nand 16.6% FLOPs employed compared with that of SSCNet.","url_abs":"http://arxiv.org/abs/1903.00620v2","url_pdf":"http://arxiv.org/pdf/1903.00620v2.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-semantic-scene-completion","task_name":"3D Semantic Scene Completion"},{"task_slug":"scene-labeling","task_name":"Scene Labeling"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-semantic-scene-completion-on-nyuv2","task":"3D Semantic Scene Completion","dataset":"NYUv2","model":"DDRNet","rank_in_archive_order":20,"of":28,"metrics":{"mIoU":"30.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.00620","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}