{"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/dense-extreme-inception-network-for-edge","title":"Dense Extreme Inception Network for Edge Detection","arxiv_id":"2112.02250","date":"2021-12-04","proceeding":null,"authors":["Xavier Soria","Angel Sappa","Patricio Humanante","Arash Akbarinia"],"abstract":"<<<This is a pre-acceptance version, please, go through Pattern Recognition Journal on Sciencedirect to read the final version>>>. Edge detection is the basis of many computer vision applications. State of the art predominantly relies on deep learning with two decisive factors: dataset content and network's architecture. Most of the publicly available datasets are not curated for edge detection tasks. Here, we offer a solution to this constraint. First, we argue that edges, contours and boundaries, despite their overlaps, are three distinct visual features requiring separate benchmark datasets. To this end, we present a new dataset of edges. Second, we propose a novel architecture, termed Dense Extreme Inception Network for Edge Detection (DexiNed), that can be trained from scratch without any pre-trained weights. DexiNed outperforms other algorithms in the presented dataset. It also generalizes well to other datasets without any fine-tuning. The higher quality of DexiNed is also perceptually evident thanks to the sharper and finer edges it outputs.","url_abs":"https://arxiv.org/abs/2112.02250v2","url_pdf":"https://arxiv.org/pdf/2112.02250v2.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":"dense-extreme-inception-network-for-edge","repo_url":"https://github.com/xavysp/DexiNed","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"edge-detection","task_name":"Edge Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/edge-detection-on-biped-1","task":"Edge Detection","dataset":"BIPED","model":"DexiNed","rank_in_archive_order":2,"of":6,"metrics":{"Number of parameters (M)":"35M","ODS":"0.895"},"uses_additional_data":false},{"leaderboard":"/sota/edge-detection-on-mdbd","task":"Edge Detection","dataset":"MDBD","model":"DexiNed-a","rank_in_archive_order":1,"of":6,"metrics":{"ODS":"0.894"},"uses_additional_data":false},{"leaderboard":"/sota/edge-detection-on-mdbd","task":"Edge Detection","dataset":"MDBD","model":"DexiNed-f","rank_in_archive_order":2,"of":6,"metrics":{"ODS":"0.891"},"uses_additional_data":false},{"leaderboard":"/sota/edge-detection-on-uded","task":"Edge Detection","dataset":"UDED","model":"DexiNed","rank_in_archive_order":4,"of":5,"metrics":{"ODS":"0.815"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2112.02250","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}