{"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/weakly-supervised-cascaded-convolutional","title":"Weakly Supervised Cascaded Convolutional Networks","arxiv_id":"1611.08258","date":"2016-11-24","proceeding":"CVPR 2017 7","authors":["Ali Diba","Vivek Sharma","Ali Pazandeh","Hamed Pirsiavash","Luc van Gool"],"abstract":"Object detection is a challenging task in visual understanding domain, and\neven more so if the supervision is to be weak. Recently, few efforts to handle\nthe task without expensive human annotations is established by promising deep\nneural network. A new architecture of cascaded networks is proposed to learn a\nconvolutional neural network (CNN) under such conditions. We introduce two such\narchitectures, with either two cascade stages or three which are trained in an\nend-to-end pipeline. The first stage of both architectures extracts best\ncandidate of class specific region proposals by training a fully convolutional\nnetwork. In the case of the three stage architecture, the middle stage provides\nobject segmentation, using the output of the activation maps of first stage.\nThe final stage of both architectures is a part of a convolutional neural\nnetwork that performs multiple instance learning on proposals extracted in the\nprevious stage(s). Our experiments on the PASCAL VOC 2007, 2010, 2012 and large\nscale object datasets, ILSVRC 2013, 2014 datasets show improvements in the\nareas of weakly-supervised object detection, classification and localization.","url_abs":"http://arxiv.org/abs/1611.08258v1","url_pdf":"http://arxiv.org/pdf/1611.08258v1.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":[],"tasks":[{"task_slug":"multiple-instance-learning","task_name":"Multiple Instance Learning"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"weakly-supervised-object-detection","task_name":"Weakly Supervised Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/weakly-supervised-object-detection-on-coco-2","task":"Weakly Supervised Object Detection","dataset":"COCO test-dev","model":"WCCN","rank_in_archive_order":3,"of":4,"metrics":{"AP50":"12.3"},"uses_additional_data":false},{"leaderboard":"/sota/weakly-supervised-object-detection-on","task":"Weakly Supervised Object Detection","dataset":"ImageNet","model":"WCCN","rank_in_archive_order":2,"of":4,"metrics":{"MAP":"16.3"},"uses_additional_data":false},{"leaderboard":"/sota/weakly-supervised-object-detection-on-pascal-1","task":"Weakly Supervised Object Detection","dataset":"PASCAL VOC 2007","model":"WCCN","rank_in_archive_order":33,"of":41,"metrics":{"MAP":"42.8"},"uses_additional_data":false},{"leaderboard":"/sota/weakly-supervised-object-detection-on-pascal","task":"Weakly Supervised Object Detection","dataset":"PASCAL VOC 2012 test","model":"WCCN","rank_in_archive_order":27,"of":32,"metrics":{"MAP":"37.9"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1611.08258","atlas_url":"https://app.syntology.ai/?focus=1611.08258","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}