{"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/craft-objects-from-images","title":"CRAFT Objects from Images","arxiv_id":"1604.03239","date":"2016-04-12","proceeding":"CVPR 2016 6","authors":["Bin Yang","Junjie Yan","Zhen Lei","Stan Z. Li"],"abstract":"Object detection is a fundamental problem in image understanding. One popular\nsolution is the R-CNN framework and its fast versions. They decompose the\nobject detection problem into two cascaded easier tasks: 1) generating object\nproposals from images, 2) classifying proposals into various object categories.\nDespite that we are handling with two relatively easier tasks, they are not\nsolved perfectly and there's still room for improvement. In this paper, we push\nthe \"divide and conquer\" solution even further by dividing each task into two\nsub-tasks. We call the proposed method \"CRAFT\" (Cascade Region-proposal-network\nAnd FasT-rcnn), which tackles each task with a carefully designed network\ncascade. We show that the cascade structure helps in both tasks: in proposal\ngeneration, it provides more compact and better localized object proposals; in\nobject classification, it reduces false positives (mainly between ambiguous\ncategories) by capturing both inter- and intra-category variances. CRAFT\nachieves consistent and considerable improvement over the state-of-the-art on\nobject detection benchmarks like PASCAL VOC 07/12 and ILSVRC.","url_abs":"http://arxiv.org/abs/1604.03239v1","url_pdf":"http://arxiv.org/pdf/1604.03239v1.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":"craft-objects-from-images","repo_url":"https://github.com/byangderek/CRAFT","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"region-proposal","task_name":"Region Proposal"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"r-cnn","method_name":"R-CNN"},{"method_slug":"svm","method_name":"SVM"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}