{"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/deformable-part-descriptors-for-fine-grained","title":"Deformable Part Descriptors for Fine-grained Recognition and Attribute Prediction","arxiv_id":null,"date":"2013-12-01","proceeding":"ICCV 2013 12","authors":["Ning Zhang","Ryan Farrell","Forrest Iandola","Trevor Darrell"],"abstract":"Recognizing objects in fine-grained domains can be\r\nextremely challenging due to the subtle differences between subcategories. Discriminative markings are often\r\nhighly localized, leading traditional object recognition approaches to struggle with the large pose variation often\r\npresent in these domains. Pose-normalization seeks to align\r\ntraining exemplars, either piecewise by part or globally\r\nfor the whole object, effectively factoring out differences\r\nin pose and in viewing angle. Prior approaches relied\r\non computationally-expensive filter ensembles for part localization and required extensive supervision. This paper proposes two pose-normalized descriptors based on\r\ncomputationally-efficient deformable part models. The\r\nfirst leverages the semantics inherent in strongly-supervised\r\nDPM parts. The second exploits weak semantic annotations to learn cross-component correspondences, computing pose-normalized descriptors from the latent parts of\r\na weakly-supervised DPM. These representations enable\r\npooling across pose and viewpoint, in turn facilitating tasks\r\nsuch as fine-grained recognition and attribute prediction.\r\nExperiments conducted on the Caltech-UCSD Birds 200\r\ndataset and Berkeley Human Attribute dataset demonstrate\r\nsignificant improvements over state-of-art algorithms.","url_abs":"https://openaccess.thecvf.com/content_iccv_2013/papers/Zhang_Deformable_Part_Descriptors_2013_ICCV_paper.pdf","url_pdf":"https://openaccess.thecvf.com/content_iccv_2013/papers/Zhang_Deformable_Part_Descriptors_2013_ICCV_paper.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":"attribute","task_name":"Attribute"},{"task_slug":"fine-grained-image-classification","task_name":"Fine-Grained Image Classification"},{"task_slug":"object-recognition","task_name":"Object Recognition"}],"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":"Deformable Part Descriptors","rank_in_archive_order":30,"of":30,"metrics":{"Accuracy":"50.98"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}