{"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/recognizing-partial-biometric-patterns","title":"Recognizing Partial Biometric Patterns","arxiv_id":"1810.07399","date":"2018-10-17","proceeding":null,"authors":["Lingxiao He","Zhenan Sun","Yuhao Zhu","Yunbo Wang"],"abstract":"Biometric recognition on partial captured targets is challenging, where only\nseveral partial observations of objects are available for matching. In this\narea, deep learning based methods are widely applied to match these partial\ncaptured objects caused by occlusions, variations of postures or just partial\nout of view in person re-identification and partial face recognition. However,\nmost current methods are not able to identify an individual in case that some\nparts of the object are not obtainable, while the rest are specialized to\ncertain constrained scenarios. To this end, we propose a robust general\nframework for arbitrary biometric matching scenarios without the limitations of\nalignment as well as the size of inputs. We introduce a feature post-processing\nstep to handle the feature maps from FCN and a dictionary learning based\nSpatial Feature Reconstruction (SFR) to match different sized feature maps in\nthis work. Moreover, the batch hard triplet loss function is applied to\noptimize the model. The applicability and effectiveness of the proposed method\nare demonstrated by the results from experiments on three person\nre-identification datasets (Market1501, CUHK03, DukeMTMC-reID), two partial\nperson datasets (Partial REID and Partial iLIDS) and two partial face datasets\n(CASIA-NIR-Distance and Partial LFW), on which state-of-the-art performance is\nensured in comparison with several state-of-the-art approaches. The code is\nreleased online and can be found on the website:\nhttps://github.com/lingxiao-he/Partial-Person-ReID.","url_abs":"http://arxiv.org/abs/1810.07399v1","url_pdf":"http://arxiv.org/pdf/1810.07399v1.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":"recognizing-partial-biometric-patterns","repo_url":"https://github.com/lingxiao-he/Partial-Person-ReID","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"dictionary-learning","task_name":"Dictionary Learning"},{"task_slug":"face-recognition","task_name":"Face Recognition"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"},{"task_slug":null,"task_name":"Triplet"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"fcn","method_name":"FCN"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"triplet-loss","method_name":"Triplet Loss"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1810.07399","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}