{"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/deep-projective-3d-semantic-segmentation","title":"Deep Projective 3D Semantic Segmentation","arxiv_id":"1705.03428","date":"2017-05-09","proceeding":null,"authors":["Felix Järemo Lawin","Martin Danelljan","Patrik Tosteberg","Goutam Bhat","Fahad Shahbaz Khan","Michael Felsberg"],"abstract":"Semantic segmentation of 3D point clouds is a challenging problem with\nnumerous real-world applications. While deep learning has revolutionized the\nfield of image semantic segmentation, its impact on point cloud data has been\nlimited so far. Recent attempts, based on 3D deep learning approaches\n(3D-CNNs), have achieved below-expected results. Such methods require\nvoxelizations of the underlying point cloud data, leading to decreased spatial\nresolution and increased memory consumption. Additionally, 3D-CNNs greatly\nsuffer from the limited availability of annotated datasets.\n  In this paper, we propose an alternative framework that avoids the\nlimitations of 3D-CNNs. Instead of directly solving the problem in 3D, we first\nproject the point cloud onto a set of synthetic 2D-images. These images are\nthen used as input to a 2D-CNN, designed for semantic segmentation. Finally,\nthe obtained prediction scores are re-projected to the point cloud to obtain\nthe segmentation results. We further investigate the impact of multiple\nmodalities, such as color, depth and surface normals, in a multi-stream network\narchitecture. Experiments are performed on the recent Semantic3D dataset. Our\napproach sets a new state-of-the-art by achieving a relative gain of 7.9 %,\ncompared to the previous best approach.","url_abs":"http://arxiv.org/abs/1705.03428v1","url_pdf":"http://arxiv.org/pdf/1705.03428v1.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":"deep-projective-3d-semantic-segmentation","repo_url":"https://github.com/TiagoCortinhal/SalsaNext","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-semantic3d","task":"Semantic Segmentation","dataset":"Semantic3D","model":"DeePr3SS","rank_in_archive_order":15,"of":17,"metrics":{"mIoU":"58.5%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.03428","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}