{"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/do-convnets-learn-correspondence","title":"Do Convnets Learn Correspondence?","arxiv_id":"1411.1091","date":"2014-11-04","proceeding":"NeurIPS 2014 12","authors":["Jonathan Long","Ning Zhang","Trevor Darrell"],"abstract":"Convolutional neural nets (convnets) trained from massive labeled datasets\nhave substantially improved the state-of-the-art in image classification and\nobject detection. However, visual understanding requires establishing\ncorrespondence on a finer level than object category. Given their large pooling\nregions and training from whole-image labels, it is not clear that convnets\nderive their success from an accurate correspondence model which could be used\nfor precise localization. In this paper, we study the effectiveness of convnet\nactivation features for tasks requiring correspondence. We present evidence\nthat convnet features localize at a much finer scale than their receptive field\nsizes, that they can be used to perform intraclass alignment as well as\nconventional hand-engineered features, and that they outperform conventional\nfeatures in keypoint prediction on objects from PASCAL VOC 2011.","url_abs":"http://arxiv.org/abs/1411.1091v1","url_pdf":"http://arxiv.org/pdf/1411.1091v1.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":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"keypoint-detection","task_name":"Keypoint Detection"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/keypoint-detection-on-pascal3d","task":"Keypoint Detection","dataset":"Pascal3D+","model":"ConvNet","rank_in_archive_order":4,"of":4,"metrics":{"Mean PCK":"48.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1411.1091","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}