{"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/bridging-saliency-detection-to-weakly","title":"Bridging Saliency Detection to Weakly Supervised Object Detection Based on Self-paced Curriculum Learning","arxiv_id":"1703.01290","date":"2017-03-03","proceeding":null,"authors":["Dingwen Zhang","Deyu Meng","Long Zhao","Junwei Han"],"abstract":"Weakly-supervised object detection (WOD) is a challenging problems in\ncomputer vision. The key problem is to simultaneously infer the exact object\nlocations in the training images and train the object detectors, given only the\ntraining images with weak image-level labels. Intuitively, by simulating the\nselective attention mechanism of human visual system, saliency detection\ntechnique can select attractive objects in scenes and thus is a potential way\nto provide useful priors for WOD. However, the way to adopt saliency detection\nin WOD is not trivial since the detected saliency region might be possibly\nhighly ambiguous in complex cases. To this end, this paper first\ncomprehensively analyzes the challenges in applying saliency detection to WOD.\nThen, we make one of the earliest efforts to bridge saliency detection to WOD\nvia the self-paced curriculum learning, which can guide the learning procedure\nto gradually achieve faithful knowledge of multi-class objects from easy to\nhard. The experimental results demonstrate that the proposed approach can\nsuccessfully bridge saliency detection and WOD tasks and achieve the\nstate-of-the-art object detection results under the weak supervision.","url_abs":"http://arxiv.org/abs/1703.01290v1","url_pdf":"http://arxiv.org/pdf/1703.01290v1.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":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"saliency-detection","task_name":"Saliency Detection"},{"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-pascal-1","task":"Weakly Supervised Object Detection","dataset":"PASCAL VOC 2007","model":"Self-paced curriculum learning","rank_in_archive_order":40,"of":41,"metrics":{"MAP":"31.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.01290","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}