{"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/cnn-based-cascaded-multi-task-learning-of","title":"CNN-based Cascaded Multi-task Learning of High-level Prior and Density Estimation for Crowd Counting","arxiv_id":"1707.09605","date":"2017-07-30","proceeding":null,"authors":["Vishwanath A. Sindagi","Vishal M. Patel"],"abstract":"Estimating crowd count in densely crowded scenes is an extremely challenging\ntask due to non-uniform scale variations. In this paper, we propose a novel\nend-to-end cascaded network of CNNs to jointly learn crowd count classification\nand density map estimation. Classifying crowd count into various groups is\ntantamount to coarsely estimating the total count in the image thereby\nincorporating a high-level prior into the density estimation network. This\nenables the layers in the network to learn globally relevant discriminative\nfeatures which aid in estimating highly refined density maps with lower count\nerror. The joint training is performed in an end-to-end fashion. Extensive\nexperiments on highly challenging publicly available datasets show that the\nproposed method achieves lower count error and better quality density maps as\ncompared to the recent state-of-the-art methods.","url_abs":"http://arxiv.org/abs/1707.09605v2","url_pdf":"http://arxiv.org/pdf/1707.09605v2.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":"cnn-based-cascaded-multi-task-learning-of","repo_url":"https://github.com/svishwa/crowdcount-cascaded-mtl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"crowd-counting","task_name":"Crowd Counting"},{"task_slug":"density-estimation","task_name":"Density Estimation"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/crowd-counting-on-shanghaitech-a","task":"Crowd Counting","dataset":"ShanghaiTech A","model":"Cascaded-MTL","rank_in_archive_order":33,"of":35,"metrics":{"MAE":"101.3","MSE":"152.4"},"uses_additional_data":false},{"leaderboard":"/sota/crowd-counting-on-shanghaitech-b","task":"Crowd Counting","dataset":"ShanghaiTech B","model":"Cascaded-MTL","rank_in_archive_order":28,"of":32,"metrics":{"MAE":"20"},"uses_additional_data":false},{"leaderboard":"/sota/crowd-counting-on-ucf-cc-50","task":"Crowd Counting","dataset":"UCF CC 50","model":"Cascaded-MTL","rank_in_archive_order":18,"of":22,"metrics":{"MAE":"322.8"},"uses_additional_data":false},{"leaderboard":"/sota/crowd-counting-on-ucf-qnrf","task":"Crowd Counting","dataset":"UCF-QNRF","model":"Cascaded-MTL","rank_in_archive_order":20,"of":23,"metrics":{"MAE":"252"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.09605","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}