{"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/improved-few-shot-visual-classification","title":"Improved Few-Shot Visual Classification","arxiv_id":"1912.03432","date":"2019-12-07","proceeding":"CVPR 2020 6","authors":["Peyman Bateni","Raghav Goyal","Vaden Masrani","Frank Wood","Leonid Sigal"],"abstract":"Few-shot learning is a fundamental task in computer vision that carries the promise of alleviating the need for exhaustively labeled data. Most few-shot learning approaches to date have focused on progressively more complex neural feature extractors and classifier adaptation strategies, as well as the refinement of the task definition itself. In this paper, we explore the hypothesis that a simple class-covariance-based distance metric, namely the Mahalanobis distance, adopted into a state of the art few-shot learning approach (CNAPS) can, in and of itself, lead to a significant performance improvement. We also discover that it is possible to learn adaptive feature extractors that allow useful estimation of the high dimensional feature covariances required by this metric from surprisingly few samples. The result of our work is a new \"Simple CNAPS\" architecture which has up to 9.2% fewer trainable parameters than CNAPS and performs up to 6.1% better than state of the art on the standard few-shot image classification benchmark dataset.","url_abs":"https://arxiv.org/abs/1912.03432v3","url_pdf":"https://arxiv.org/pdf/1912.03432v3.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":"improved-few-shot-visual-classification","repo_url":"https://github.com/plai-group/simple-cnaps","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"improved-few-shot-visual-classification","repo_url":"https://github.com/peymanbateni/simple-cnaps","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"few-shot-image-classification","task_name":"Few-Shot Image Classification"},{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"object-recognition","task_name":"Object Recognition"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/few-shot-image-classification-on-meta-dataset","task":"Few-Shot Image Classification","dataset":"Meta-Dataset","model":"Simple CNAPS","rank_in_archive_order":12,"of":22,"metrics":{"Accuracy":"69.86"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-meta-dataset-1","task":"Few-Shot Image Classification","dataset":"Meta-Dataset Rank","model":"Simple CNAPS","rank_in_archive_order":3,"of":13,"metrics":{"Mean Rank":"3.45"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-12","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (1-shot)","model":"Simple CNAPS + FETI","rank_in_archive_order":2,"of":14,"metrics":{"Accuracy":"63.5"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-mini-12","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (1-shot)","model":"Simple CNAPS","rank_in_archive_order":6,"of":14,"metrics":{"Accuracy":"37.1"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-13","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (5-shot)","model":"Simple CNAPS + FETI","rank_in_archive_order":2,"of":14,"metrics":{"Accuracy":"83.1"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-mini-13","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 10-way (5-shot)","model":"Simple CNAPS","rank_in_archive_order":5,"of":14,"metrics":{"Accuracy":"56.7"},"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":"Simple CNAPS + FETI","rank_in_archive_order":19,"of":105,"metrics":{"Accuracy":"77.4"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-mini-2","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 5-way (1-shot)","model":"Simple CNAPS","rank_in_archive_order":90,"of":105,"metrics":{"Accuracy":"53.2"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-mini-3","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 5-way (5-shot)","model":"Simple CNAPS + FETI","rank_in_archive_order":11,"of":95,"metrics":{"Accuracy":"90.3"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-mini-3","task":"Few-Shot Image Classification","dataset":"Mini-Imagenet 5-way (5-shot)","model":"Simple CNAPS","rank_in_archive_order":81,"of":95,"metrics":{"Accuracy":"70.8"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-2","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (1-shot)","model":"Simple CNAPS + FETI","rank_in_archive_order":2,"of":13,"metrics":{"Accuracy":"57.1"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-2","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (1-shot)","model":"Simple CNAPS","rank_in_archive_order":4,"of":13,"metrics":{"Accuracy":"48.1"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-3","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (5-shot)","model":"Simple CNAPS + FETI","rank_in_archive_order":2,"of":13,"metrics":{"Accuracy":"78.5"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-3","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 10-way (5-shot)","model":"Simple CNAPS","rank_in_archive_order":4,"of":13,"metrics":{"Accuracy":"70.2"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 5-way (1-shot)","model":"Simple CNAPS + FETI","rank_in_archive_order":29,"of":49,"metrics":{"Accuracy":"71.4"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-tiered","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 5-way (1-shot)","model":"Simple CNAPS","rank_in_archive_order":45,"of":49,"metrics":{"Accuracy":"63.0"},"uses_additional_data":false},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-1","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 5-way (5-shot)","model":"Simple CNAPS + FETI","rank_in_archive_order":29,"of":51,"metrics":{"Accuracy":"86.0"},"uses_additional_data":true},{"leaderboard":"/sota/few-shot-image-classification-on-tiered-1","task":"Few-Shot Image Classification","dataset":"Tiered ImageNet 5-way (5-shot)","model":"Simple CNAPS","rank_in_archive_order":45,"of":51,"metrics":{"Accuracy":"80.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1912.03432","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}