{"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/semantic-point-completion-network-for-3d","title":"Semantic Point Completion Network for 3D Semantic Scene Completion","arxiv_id":null,"date":"2020-08-29","proceeding":"ECAI 2020 8","authors":["Min Zhong","Gang Zeng"],"abstract":"Semantic scene completion (SSC) is composed of scene\r\ncompletion (SC) and semantic segmentation. Most of the existing\r\nmethods carry out SSC in a regular 3D grid space, where 3D CNNs\r\ncause unnecessary computational cost on empty voxels. In this work,\r\na Semantic Point Completion Network (SPCNet) is proposed to\r\naddress SSC in the point cloud space. Specifically, SPCNet is an\r\nEncoder-decoder architecture, in which an Observed Point Encoder\r\nis applied to extract the features of observed points, and an Observed\r\nto Occluded Point Decoder is responsible for mapping the features\r\nto the occluded points. Based on the SPCNet, we further introduce\r\nan Image-point Fused Semantic Point Completion Network (IPFSPCNet), which aims to boost the performance of SSC by combining\r\nthe texture with geometry information. Evaluations are conducted on\r\ntwo public datasets. Experimental results show that our method can\r\naddress the SC problem in the point cloud space. Compared to stateof-the-art approaches, our method can achieve satisfying results on\r\nthe SSC task.","url_abs":"https://ecai2020.eu/papers/123_paper.pdf","url_pdf":"https://ecai2020.eu/papers/123_paper.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":"decoder","task_name":"Decoder"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-semantic-scene-completion-on-nyuv2","task":"3D Semantic Scene Completion","dataset":"NYUv2","model":"IPF-SPCNet: Semantic point completion network for 3D\nsemantic scene completion","rank_in_archive_order":9,"of":28,"metrics":{"mIoU":"35.1"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}