{"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/efficient-object-localization-using","title":"Efficient Object Localization Using Convolutional Networks","arxiv_id":"1411.4280","date":"2014-11-16","proceeding":"CVPR 2015 6","authors":["Jonathan Tompson","Ross Goroshin","Arjun Jain","Yann Lecun","Christopher Bregler"],"abstract":"Recent state-of-the-art performance on human-body pose estimation has been\nachieved with Deep Convolutional Networks (ConvNets). Traditional ConvNet\narchitectures include pooling and sub-sampling layers which reduce\ncomputational requirements, introduce invariance and prevent over-training.\nThese benefits of pooling come at the cost of reduced localization accuracy. We\nintroduce a novel architecture which includes an efficient `position\nrefinement' model that is trained to estimate the joint offset location within\na small region of the image. This refinement model is jointly trained in\ncascade with a state-of-the-art ConvNet model to achieve improved accuracy in\nhuman joint location estimation. We show that the variance of our detector\napproaches the variance of human annotations on the FLIC dataset and\noutperforms all existing approaches on the MPII-human-pose dataset.","url_abs":"http://arxiv.org/abs/1411.4280v3","url_pdf":"http://arxiv.org/pdf/1411.4280v3.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":"efficient-object-localization-using","repo_url":"https://github.com/cmu-enyac/Renofeation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"efficient-object-localization-using","repo_url":"https://github.com/yukitsuji/chainer_function","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-localization","task_name":"Object Localization"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":null,"task_name":"Position"}],"methods":[{"method_slug":"spatialdropout","method_name":"SpatialDropout"}],"datasets_introduced":[],"methods_introduced":[{"slug":"spatialdropout","name":"SpatialDropout","full_name":"SpatialDropout"}],"results":[{"leaderboard":"/sota/pose-estimation-on-mpii-human-pose","task":"Pose Estimation","dataset":"MPII Human Pose","model":"Tompson et al.","rank_in_archive_order":42,"of":46,"metrics":{"PCKh-0.5":"82.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1411.4280","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}