{"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/real-time-self-adaptive-deep-stereo","title":"Real-time self-adaptive deep stereo","arxiv_id":"1810.05424","date":"2018-10-12","proceeding":"CVPR 2019 6","authors":["Alessio Tonioni","Fabio Tosi","Matteo Poggi","Stefano Mattoccia","Luigi Di Stefano"],"abstract":"Deep convolutional neural networks trained end-to-end are the\nstate-of-the-art methods to regress dense disparity maps from stereo pairs.\nThese models, however, suffer from a notable decrease in accuracy when exposed\nto scenarios significantly different from the training set, e.g., real vs\nsynthetic images, etc.). We argue that it is extremely unlikely to gather\nenough samples to achieve effective training/tuning in any target domain, thus\nmaking this setup impractical for many applications. Instead, we propose to\nperform unsupervised and continuous online adaptation of a deep stereo network,\nwhich allows for preserving its accuracy in any environment. However, this\nstrategy is extremely computationally demanding and thus prevents real-time\ninference. We address this issue introducing a new lightweight, yet effective,\ndeep stereo architecture, Modularly ADaptive Network (MADNet) and developing a\nModular ADaptation (MAD) algorithm, which independently trains sub-portions of\nthe network. By deploying MADNet together with MAD we introduce the first\nreal-time self-adaptive deep stereo system enabling competitive performance on\nheterogeneous datasets.","url_abs":"http://arxiv.org/abs/1810.05424v2","url_pdf":"http://arxiv.org/pdf/1810.05424v2.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":"real-time-self-adaptive-deep-stereo","repo_url":"https://github.com/CVLAB-Unibo/Real-time-self-adaptive-deep-stereo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"stereo-depth-estimation","task_name":"Stereo Depth Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1810.05424","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}