{"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/aga-attribute-guided-augmentation","title":"AGA: Attribute Guided Augmentation","arxiv_id":"1612.02559","date":"2016-12-08","proceeding":null,"authors":["Mandar Dixit","Roland Kwitt","Marc Niethammer","Nuno Vasconcelos"],"abstract":"We consider the problem of data augmentation, i.e., generating artificial\nsamples to extend a given corpus of training data. Specifically, we propose\nattributed-guided augmentation (AGA) which learns a mapping that allows to\nsynthesize data such that an attribute of a synthesized sample is at a desired\nvalue or strength. This is particularly interesting in situations where little\ndata with no attribute annotation is available for learning, but we have access\nto a large external corpus of heavily annotated samples. While prior works\nprimarily augment in the space of images, we propose to perform augmentation in\nfeature space instead. We implement our approach as a deep encoder-decoder\narchitecture that learns the synthesis function in an end-to-end manner. We\ndemonstrate the utility of our approach on the problems of (1) one-shot object\nrecognition in a transfer-learning setting where we have no prior knowledge of\nthe new classes, as well as (2) object-based one-shot scene recognition. As\nexternal data, we leverage 3D depth and pose information from the SUN RGB-D\ndataset. Our experiments show that attribute-guided augmentation of high-level\nCNN features considerably improves one-shot recognition performance on both\nproblems.","url_abs":"http://arxiv.org/abs/1612.02559v2","url_pdf":"http://arxiv.org/pdf/1612.02559v2.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":"aga-attribute-guided-augmentation","repo_url":"https://github.com/rkwitt/GuidedAugmentation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"torch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"object-recognition","task_name":"Object Recognition"},{"task_slug":"scene-recognition","task_name":"Scene Recognition"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.02559","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}