{"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/omnidepth-dense-depth-estimation-for-indoors","title":"OmniDepth: Dense Depth Estimation for Indoors Spherical Panoramas","arxiv_id":"1807.09620","date":"2018-07-25","proceeding":"ECCV 2018 9","authors":["Nikolaos Zioulis","Antonis Karakottas","Dimitrios Zarpalas","Petros Daras"],"abstract":"Recent work on depth estimation up to now has only focused on projective\nimages ignoring 360 content which is now increasingly and more easily produced.\nWe show that monocular depth estimation models trained on traditional images\nproduce sub-optimal results on omnidirectional images, showcasing the need for\ntraining directly on 360 datasets, which however, are hard to acquire. In this\nwork, we circumvent the challenges associated with acquiring high quality 360\ndatasets with ground truth depth annotations, by re-using recently released\nlarge scale 3D datasets and re-purposing them to 360 via rendering. This\ndataset, which is considerably larger than similar projective datasets, is\npublicly offered to the community to enable future research in this direction.\nWe use this dataset to learn in an end-to-end fashion the task of depth\nestimation from 360 images. We show promising results in our synthesized data\nas well as in unseen realistic images.","url_abs":"http://arxiv.org/abs/1807.09620v1","url_pdf":"http://arxiv.org/pdf/1807.09620v1.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":"omnidepth-dense-depth-estimation-for-indoors","repo_url":"https://github.com/VCL3D/SphericalViewSynthesis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-2-Clause"}}],"tasks":[{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"monocular-depth-estimation","task_name":"Monocular Depth Estimation"}],"methods":[],"datasets_introduced":[{"slug":"3d60","name":"3D60","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/depth-estimation-on-stanford2d3d-panoramic","task":"Depth Estimation","dataset":"Stanford2D3D Panoramic","model":"OmniDepth","rank_in_archive_order":18,"of":18,"metrics":{"RMSE":"0.6152","absolute relative error":"0.1996 "},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.09620","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}