{"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/delta-encoder-an-effective-sample-synthesis","title":"Delta-encoder: an effective sample synthesis method for few-shot object recognition","arxiv_id":"1806.04734","date":"2018-06-12","proceeding":"NeurIPS 2018 12","authors":["Eli Schwartz","Leonid Karlinsky","Joseph Shtok","Sivan Harary","Mattias Marder","Rogerio Feris","Abhishek Kumar","Raja Giryes","Alex M. Bronstein"],"abstract":"Learning to classify new categories based on just one or a few examples is a\nlong-standing challenge in modern computer vision. In this work, we proposes a\nsimple yet effective method for few-shot (and one-shot) object recognition. Our\napproach is based on a modified auto-encoder, denoted Delta-encoder, that\nlearns to synthesize new samples for an unseen category just by seeing few\nexamples from it. The synthesized samples are then used to train a classifier.\nThe proposed approach learns to both extract transferable intra-class\ndeformations, or \"deltas\", between same-class pairs of training examples, and\nto apply those deltas to the few provided examples of a novel class (unseen\nduring training) in order to efficiently synthesize samples from that new\nclass. The proposed method improves over the state-of-the-art in one-shot\nobject-recognition and compares favorably in the few-shot case. Upon acceptance\ncode will be made available.","url_abs":"http://arxiv.org/abs/1806.04734v3","url_pdf":"http://arxiv.org/pdf/1806.04734v3.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":"delta-encoder-an-effective-sample-synthesis","repo_url":"https://github.com/EliSchwartz/DeltaEncoder","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"few-shot-image-classification","task_name":"Few-Shot Image Classification"},{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"object-recognition","task_name":"Object Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/few-shot-image-classification-on-cifar100-5","task":"Few-Shot Image Classification","dataset":"CIFAR100 5-way (1-shot)","model":"Delta-encoder","rank_in_archive_order":2,"of":2,"metrics":{"Accuracy":"66.7"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-cub-200-5-1","task":"Few-Shot Image Classification","dataset":"CUB 200 5-way 1-shot","model":"Delta-encoder","rank_in_archive_order":29,"of":36,"metrics":{"Accuracy":"69.8"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-caltech-256","task":"Few-Shot Image Classification","dataset":"Caltech-256 5-way (1-shot)","model":"Delta-encoder","rank_in_archive_order":2,"of":3,"metrics":{"Accuracy":"73.2"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-2","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 5-way (1-shot)","model":"Delta-encoder","rank_in_archive_order":74,"of":105,"metrics":{"Accuracy":"59.9"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.04734","atlas_url":"https://app.syntology.ai/?focus=1806.04734","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}