{"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/view-volume-network-for-semantic-scene","title":"View-volume Network for Semantic Scene Completion from a Single Depth Image","arxiv_id":"1806.05361","date":"2018-06-14","proceeding":null,"authors":["Yu-Xiao Guo","Xin Tong"],"abstract":"We introduce a View-Volume convolutional neural network (VVNet) for inferring\nthe occupancy and semantic labels of a volumetric 3D scene from a single depth\nimage. The VVNet concatenates a 2D view CNN and a 3D volume CNN with a\ndifferentiable projection layer. Given a single RGBD image, our method extracts\nthe detailed geometric features from the input depth image with a 2D view CNN\nand then projects the features into a 3D volume according to the input depth\nmap via a projection layer. After that, we learn the 3D context information of\nthe scene with a 3D volume CNN for computing the result volumetric occupancy\nand semantic labels. With combined 2D and 3D representations, the VVNet\nefficiently reduces the computational cost, enables feature extraction from\nmulti-channel high resolution inputs, and thus significantly improves the\nresult accuracy. We validate our method and demonstrate its efficiency and\neffectiveness on both synthetic SUNCG and real NYU dataset.","url_abs":"http://arxiv.org/abs/1806.05361v1","url_pdf":"http://arxiv.org/pdf/1806.05361v1.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"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-semantic-scene-completion-on-nyuv2","task":"3D Semantic Scene Completion","dataset":"NYUv2","model":"VVNet","rank_in_archive_order":21,"of":28,"metrics":{"mIoU":"29.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.05361","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}