{"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/pairwise-confusion-for-fine-grained-visual","title":"Pairwise Confusion for Fine-Grained Visual Classification","arxiv_id":"1705.08016","date":"2017-05-22","proceeding":"ECCV 2018 9","authors":["Abhimanyu Dubey","Otkrist Gupta","Pei Guo","Ramesh Raskar","Ryan Farrell","Nikhil Naik"],"abstract":"Fine-Grained Visual Classification (FGVC) datasets contain small sample\nsizes, along with significant intra-class variation and inter-class similarity.\nWhile prior work has addressed intra-class variation using localization and\nsegmentation techniques, inter-class similarity may also affect feature\nlearning and reduce classification performance. In this work, we address this\nproblem using a novel optimization procedure for the end-to-end neural network\ntraining on FGVC tasks. Our procedure, called Pairwise Confusion (PC) reduces\noverfitting by intentionally {introducing confusion} in the activations. With\nPC regularization, we obtain state-of-the-art performance on six of the most\nwidely-used FGVC datasets and demonstrate improved localization ability. {PC}\nis easy to implement, does not need excessive hyperparameter tuning during\ntraining, and does not add significant overhead during test time.","url_abs":"http://arxiv.org/abs/1705.08016v3","url_pdf":"http://arxiv.org/pdf/1705.08016v3.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":"pairwise-confusion-for-fine-grained-visual","repo_url":"https://github.com/abhimanyudubey/confusion","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"fine-grained-image-classification","task_name":"Fine-Grained Image Classification"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/fine-grained-image-classification-on-cub-200-1","task":"Fine-Grained Image Classification","dataset":"CUB-200-2011","model":"PC","rank_in_archive_order":26,"of":30,"metrics":{"Accuracy":"86.9"},"uses_additional_data":false},{"leaderboard":"/sota/fine-grained-image-classification-on-nabirds","task":"Fine-Grained Image Classification","dataset":"NABirds","model":"PC-DenseNet-161","rank_in_archive_order":28,"of":30,"metrics":{"Accuracy":"82.79%"},"uses_additional_data":false},{"leaderboard":"/sota/fine-grained-image-classification-on-oxford","task":"Fine-Grained Image Classification","dataset":"Oxford 102 Flowers","model":"PC Bilinear CNN","rank_in_archive_order":22,"of":25,"metrics":{"Accuracy":"93.65%"},"uses_additional_data":false},{"leaderboard":"/sota/fine-grained-image-classification-on-stanford","task":"Fine-Grained Image Classification","dataset":"Stanford Cars","model":"PC-DenseNet-161","rank_in_archive_order":69,"of":83,"metrics":{"Accuracy":"92.86%"},"uses_additional_data":false},{"leaderboard":"/sota/fine-grained-image-classification-on-stanford-1","task":"Fine-Grained Image Classification","dataset":"Stanford Dogs","model":"PC-DenseNet-161","rank_in_archive_order":23,"of":24,"metrics":{"Accuracy":"83.75%"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}