{"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/learning-monocular-depth-estimation-with","title":"Learning monocular depth estimation with unsupervised trinocular assumptions","arxiv_id":"1808.01606","date":"2018-08-05","proceeding":null,"authors":["Matteo Poggi","Fabio Tosi","Stefano Mattoccia"],"abstract":"Obtaining accurate depth measurements out of a single image represents a\nfascinating solution to 3D sensing. CNNs led to considerable improvements in\nthis field, and recent trends replaced the need for ground-truth labels with\ngeometry-guided image reconstruction signals enabling unsupervised training.\nCurrently, for this purpose, state-of-the-art techniques rely on images\nacquired with a binocular stereo rig to predict inverse depth (i.e., disparity)\naccording to the aforementioned supervision principle. However, these methods\nsuffer from well-known problems near occlusions, left image border, etc\ninherited from the stereo setup. Therefore, in this paper, we tackle these\nissues by moving to a trinocular domain for training. Assuming the central\nimage as the reference, we train a CNN to infer disparity representations\npairing such image with frames on its left and right side. This strategy allows\nobtaining depth maps not affected by typical stereo artifacts. Moreover, being\ntrinocular datasets seldom available, we introduce a novel interleaved training\nprocedure enabling to enforce the trinocular assumption outlined from current\nbinocular datasets. Exhaustive experimental results on the KITTI dataset\nconfirm that our proposal outperforms state-of-the-art methods for unsupervised\nmonocular depth estimation trained on binocular stereo pairs as well as any\nknown methods relying on other cues.","url_abs":"http://arxiv.org/abs/1808.01606v1","url_pdf":"http://arxiv.org/pdf/1808.01606v1.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":"learning-monocular-depth-estimation-with","repo_url":"https://github.com/mattpoggi/3net","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"image-reconstruction","task_name":"Image Reconstruction"},{"task_slug":"monocular-depth-estimation","task_name":"Monocular Depth Estimation"},{"task_slug":"unsupervised-monocular-depth-estimation","task_name":"Unsupervised Monocular Depth Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/monocular-depth-estimation-on-kitti-eigen","task":"Monocular Depth Estimation","dataset":"KITTI Eigen split","model":"3Net","rank_in_archive_order":69,"of":79,"metrics":{"absolute relative error":"0.126"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.01606","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}