{"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/signet-semantic-instance-aided-unsupervised","title":"SIGNet: Semantic Instance Aided Unsupervised 3D Geometry Perception","arxiv_id":"1812.05642","date":"2018-12-13","proceeding":"CVPR 2019 6","authors":["Yue Meng","Yongxi Lu","Aman Raj","Samuel Sunarjo","Rui Guo","Tara Javidi","Gaurav Bansal","Dinesh Bharadia"],"abstract":"Unsupervised learning for geometric perception (depth, optical flow, etc.) is\nof great interest to autonomous systems. Recent works on unsupervised learning\nhave made considerable progress on perceiving geometry; however, they usually\nignore the coherence of objects and perform poorly under scenarios with dark\nand noisy environments. In contrast, supervised learning algorithms, which are\nrobust, require large labeled geometric dataset. This paper introduces SIGNet,\na novel framework that provides robust geometry perception without requiring\ngeometrically informative labels. Specifically, SIGNet integrates semantic\ninformation to make depth and flow predictions consistent with objects and\nrobust to low lighting conditions. SIGNet is shown to improve upon the\nstate-of-the-art unsupervised learning for depth prediction by 30% (in squared\nrelative error). In particular, SIGNet improves the dynamic object class\nperformance by 39% in depth prediction and 29% in flow prediction. Our code\nwill be made available at https://github.com/mengyuest/SIGNet","url_abs":"http://arxiv.org/abs/1812.05642v2","url_pdf":"http://arxiv.org/pdf/1812.05642v2.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":"signet-semantic-instance-aided-unsupervised","repo_url":"https://github.com/mengyuest/SIGNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"signet-semantic-instance-aided-unsupervised","repo_url":"https://github.com/raunaks13/carla-SIGNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"3d-geometry-perception","task_name":"3D Geometry Perception"},{"task_slug":"3d-geometry","task_name":"3D geometry"},{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"depth-prediction","task_name":"Depth Prediction"},{"task_slug":"monocular-depth-estimation","task_name":"Monocular Depth Estimation"},{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/monocular-depth-estimation-on-kitti-eigen","task":"Monocular Depth Estimation","dataset":"KITTI Eigen split","model":"SIGNet","rank_in_archive_order":72,"of":79,"metrics":{"absolute relative error":"0.133"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.05642","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}