{"url":"/sota/few-shot-object-detection-on-ms-coco-10-shot","task":{"name":"Few-Shot Object Detection","url":"/task/few-shot-object-detection","note":null},"dataset":{"name":"MS-COCO (10-shot)","url":"/dataset/coco"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**Few-Shot Object Detection** is a computer vision task that involves detecting objects in images with limited training data. The goal is to train a model on a few examples of each object class and then use the model to detect objects in new images.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["AP"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"AP":"higher"}},"counts":{"rows":33,"rows_with_code":27,"rows_with_paper_page":33,"rows_dated":33,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"Training-free","metrics":{"AP":"36.6"},"uses_additional_data":false,"paper_date":"2025-07-03","paper":"/paper/no-time-to-train-training-free-reference","paper_url":"https://arxiv.org/abs/2507.02798v1","paper_title":"No time to train! Training-Free Reference-Based Instance Segmentation","code":"https://github.com/miquel-espinosa/no-time-to-train","n_code_links":2,"syntology":null},{"rank_in_archive_order":2,"model":"CD-ViTO","metrics":{"AP":"35.3"},"uses_additional_data":false,"paper_date":"2024-02-05","paper":"/paper/cross-domain-few-shot-object-detection-via","paper_url":"https://arxiv.org/abs/2402.03094v4","paper_title":"Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detector","code":"https://github.com/lovelyqian/CDFSOD-benchmark","n_code_links":2,"syntology":{"n_ran":9,"n_unverified":2,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"DE-ViT","metrics":{"AP":"34.0"},"uses_additional_data":false,"paper_date":"2023-09-22","paper":"/paper/detect-every-thing-with-few-examples","paper_url":"https://arxiv.org/abs/2309.12969v4","paper_title":"Detect Everything with Few Examples","code":"https://github.com/mlzxy/devit","n_code_links":1,"syntology":{"n_ran":15,"n_unverified":5,"n_samples":20,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"BIOT","metrics":{"AP":"26.3"},"uses_additional_data":false,"paper_date":"2024-05-23","paper":"/paper/balanced-id-ood-tradeoff-transfer-makes-query","paper_url":"https://www.sciencedirect.com/science/article/pii/S2667295224000400","paper_title":"Balanced ID-OOD tradeoff transfer makes query based detectors good few shot learners","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"RISF (SWIN-Large)","metrics":{"AP":"25.5"},"uses_additional_data":false,"paper_date":"2023-11-01","paper":"/paper/re-scoring-using-image-language-similarity","paper_url":"https://arxiv.org/abs/2311.00278v1","paper_title":"Re-Scoring Using Image-Language Similarity for Few-Shot Object Detection","code":"https://github.com/INFINIQ-AI1/RISF","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"DETReg-ft-full DDETR","metrics":{"AP":"25"},"uses_additional_data":false,"paper_date":"2021-06-08","paper":"/paper/detreg-unsupervised-pretraining-with-region","paper_url":"https://arxiv.org/abs/2106.04550v5","paper_title":"DETReg: Unsupervised Pretraining with Region Priors for Object Detection","code":"https://github.com/amirbar/detreg","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"imTED+ViT-B","metrics":{"AP":"22.5"},"uses_additional_data":false,"paper_date":"2022-05-19","paper":"/paper/integral-migrating-pre-trained-transformer","paper_url":"https://arxiv.org/abs/2205.09613v2","paper_title":"Integrally Migrating Pre-trained Transformer Encoder-decoders for Visual Object Detection","code":"https://github.com/yuhongtian17/spatial-transform-decoupling","n_code_links":3,"syntology":null},{"rank_in_archive_order":8,"model":"hANMCL","metrics":{"AP":"22.4"},"uses_additional_data":false,"paper_date":"2022-08-15","paper":"/paper/hierarchical-attention-network-for-few-shot","paper_url":"https://arxiv.org/abs/2208.07039v3","paper_title":"Hierarchical Attention Network for Few-Shot Object Detection via Meta-Contrastive Learning","code":"https://github.com/infinity7428/hANMCL","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"RISF (Resnet-101)","metrics":{"AP":"21.9"},"uses_additional_data":false,"paper_date":"2023-11-01","paper":"/paper/re-scoring-using-image-language-similarity","paper_url":"https://arxiv.org/abs/2311.00278v1","paper_title":"Re-Scoring Using Image-Language Similarity for Few-Shot Object Detection","code":"https://github.com/INFINIQ-AI1/RISF","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"DCFS","metrics":{"AP":"19.5"},"uses_additional_data":false,"paper_date":"2025-05-20","paper":"/paper/decoupling-classifier-for-boosting-few-shot-1","paper_url":"https://arxiv.org/abs/2505.14239v1","paper_title":"Decoupling Classifier for Boosting Few-shot Object Detection and Instance Segmentation","code":"https://github.com/gaobb/DCFS","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":3,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"CFA-DeFRCN","metrics":{"AP":"19.1"},"uses_additional_data":false,"paper_date":"2022-04-11","paper":"/paper/cfa-constraint-based-finetuning-approach-for","paper_url":"https://arxiv.org/abs/2204.05220v1","paper_title":"CFA: Constraint-based Finetuning Approach for Generalized Few-Shot Object Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"DAnA-FasterRCNN","metrics":{"AP":"18.6"},"uses_additional_data":false,"paper_date":"2021-02-24","paper":"/paper/should-i-look-at-the-head-or-the-tail-dual","paper_url":"https://arxiv.org/abs/2102.12152v3","paper_title":"Dual-Awareness Attention for Few-Shot Object Detection","code":"https://github.com/Tung-I/Dual-awareness-Attention-for-Few-shot-Object-Detection","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"DeFRCN","metrics":{"AP":"18.5"},"uses_additional_data":false,"paper_date":"2021-08-20","paper":"/paper/defrcn-decoupled-faster-r-cnn-for-few-shot","paper_url":"https://arxiv.org/abs/2108.09017v1","paper_title":"DeFRCN: Decoupled Faster R-CNN for Few-Shot Object Detection","code":"https://github.com/er-muyue/defrcn","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"Meta-DETR (Multi-Scale Feature)","metrics":{"AP":"17.8"},"uses_additional_data":false,"paper_date":"2021-03-22","paper":"/paper/meta-detr-few-shot-object-detection-via","paper_url":"https://arxiv.org/abs/2103.11731v3","paper_title":"Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation","code":"https://github.com/ZhangGongjie/Meta-DETR","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"Meta-DETR (Single-Scale Feature)","metrics":{"AP":"16.7"},"uses_additional_data":false,"paper_date":"2021-03-22","paper":"/paper/meta-detr-few-shot-object-detection-via","paper_url":"https://arxiv.org/abs/2103.11731v3","paper_title":"Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation","code":"https://github.com/ZhangGongjie/Meta-DETR","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"FSRN (RetinaNet)","metrics":{"AP":"15.8"},"uses_additional_data":false,"paper_date":"2022-10-11","paper":"/paper/towards-discriminative-and-transferable-one","paper_url":"https://arxiv.org/abs/2210.05783v1","paper_title":"Towards Discriminative and Transferable One-Stage Few-Shot Object Detectors","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":17,"model":"CME","metrics":{"AP":"15.1"},"uses_additional_data":false,"paper_date":"2021-03-08","paper":"/paper/beyond-max-margin-class-margin-equilibrium","paper_url":"https://arxiv.org/abs/2103.04612v3","paper_title":"Beyond Max-Margin: Class Margin Equilibrium for Few-shot Object Detection","code":"https://github.com/Bohao-Lee/CME","n_code_links":4,"syntology":{"n_ran":3,"n_unverified":3,"n_samples":6,"n_pointer_only_licence":6}},{"rank_in_archive_order":18,"model":"imTED+ViT-S","metrics":{"AP":"15.0"},"uses_additional_data":false,"paper_date":"2022-05-19","paper":"/paper/integral-migrating-pre-trained-transformer","paper_url":"https://arxiv.org/abs/2205.09613v2","paper_title":"Integrally Migrating Pre-trained Transformer Encoder-decoders for Visual Object Detection","code":"https://github.com/yuhongtian17/spatial-transform-decoupling","n_code_links":3,"syntology":null},{"rank_in_archive_order":19,"model":"FSDetView + PSP","metrics":{"AP":"13.4"},"uses_additional_data":false,"paper_date":"2021-09-16","paper":"/paper/few-shot-object-detection-by-attending-to-per","paper_url":"https://arxiv.org/abs/2109.07734v1","paper_title":"Few-Shot Object Detection by Attending to Per-Sample-Prototype","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"PnP-FSOD + CT","metrics":{"AP":"13.3"},"uses_additional_data":false,"paper_date":"2021-10-26","paper":"/paper/plug-and-play-few-shot-object-detection-with","paper_url":"https://arxiv.org/abs/2110.13377v2","paper_title":"Instant Response Few-shot Object Detection with Meta Strategy and Explicit Localization Inference","code":"https://github.com/16337093/PnP-FSOD","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"FSDetView","metrics":{"AP":"12.5"},"uses_additional_data":false,"paper_date":"2020-07-23","paper":"/paper/few-shot-object-detection-and-viewpoint","paper_url":"https://arxiv.org/abs/2007.12107v2","paper_title":"Few-Shot Object Detection and Viewpoint Estimation for Objects in the Wild","code":"https://github.com/YoungXIAO13/FewShotDetection","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":16,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"PnP-FSOD","metrics":{"AP":"12.5"},"uses_additional_data":false,"paper_date":"2021-10-26","paper":"/paper/plug-and-play-few-shot-object-detection-with","paper_url":"https://arxiv.org/abs/2110.13377v2","paper_title":"Instant Response Few-shot Object Detection with Meta Strategy and Explicit Localization Inference","code":"https://github.com/16337093/PnP-FSOD","n_code_links":1,"syntology":null},{"rank_in_archive_order":23,"model":"SSR-FSD","metrics":{"AP":"11.3"},"uses_additional_data":false,"paper_date":"2021-03-02","paper":"/paper/semantic-relation-reasoning-for-shot-stable","paper_url":"https://arxiv.org/abs/2103.01903v2","paper_title":"Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":24,"model":"FSOD","metrics":{"AP":"11.1"},"uses_additional_data":false,"paper_date":"2019-08-06","paper":"/paper/few-shot-object-detection-with-attention-rpn","paper_url":"https://arxiv.org/abs/1908.01998v4","paper_title":"Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector","code":"https://github.com/fanq15/Few-Shot-Object-Detection-Dataset","n_code_links":3,"syntology":{"n_ran":2,"n_unverified":8,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":25,"model":"FSCE","metrics":{"AP":"11.1"},"uses_additional_data":false,"paper_date":"2021-03-10","paper":"/paper/fsce-few-shot-object-detection-via","paper_url":"https://arxiv.org/abs/2103.05950v2","paper_title":"FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding","code":"https://github.com/megvii-research/fsce","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"FSOD(Universal-Prototype)","metrics":{"AP":"11.0"},"uses_additional_data":false,"paper_date":"2021-03-01","paper":"/paper/universal-prototype-augmentation-for-few-shot","paper_url":"https://arxiv.org/abs/2103.01077v2","paper_title":"Universal-Prototype Enhancing for Few-Shot Object Detection","code":"https://github.com/amingwu/up-fsod","n_code_links":1,"syntology":null},{"rank_in_archive_order":27,"model":"TFA(w/cos)","metrics":{"AP":"10.0"},"uses_additional_data":false,"paper_date":"2020-03-16","paper":"/paper/frustratingly-simple-few-shot-object","paper_url":"https://arxiv.org/abs/2003.06957v1","paper_title":"Frustratingly Simple Few-Shot Object Detection","code":"https://github.com/ucbdrive/few-shot-object-detection","n_code_links":5,"syntology":{"n_ran":3,"n_unverified":8,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":28,"model":"TFA(w/fc)","metrics":{"AP":"10.0"},"uses_additional_data":false,"paper_date":"2020-03-16","paper":"/paper/frustratingly-simple-few-shot-object","paper_url":"https://arxiv.org/abs/2003.06957v1","paper_title":"Frustratingly Simple Few-Shot Object Detection","code":"https://github.com/ucbdrive/few-shot-object-detection","n_code_links":5,"syntology":{"n_ran":3,"n_unverified":8,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":29,"model":"MPSR","metrics":{"AP":"9.8"},"uses_additional_data":false,"paper_date":"2020-07-18","paper":"/paper/multi-scale-positive-sample-refinement-for","paper_url":"https://arxiv.org/abs/2007.09384v1","paper_title":"Multi-Scale Positive Sample Refinement for Few-Shot Object Detection","code":"https://github.com/er-muyue/defrcn","n_code_links":4,"syntology":{"n_ran":5,"n_unverified":3,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":30,"model":"Meta R-CNN","metrics":{"AP":"8.7"},"uses_additional_data":false,"paper_date":"2019-10-01","paper":"/paper/meta-r-cnn-towards-general-solver-for-1","paper_url":"http://openaccess.thecvf.com/content_ICCV_2019/html/Yan_Meta_R-CNN_Towards_General_Solver_for_Instance-Level_Low-Shot_Learning_ICCV_2019_paper.html","paper_title":"Meta R-CNN: Towards General Solver for Instance-Level Low-Shot Learning","code":"https://github.com/yanxp/MetaR-CNN","n_code_links":1,"syntology":null},{"rank_in_archive_order":31,"model":"MetaDet","metrics":{"AP":"7.1"},"uses_additional_data":false,"paper_date":"2019-10-01","paper":"/paper/meta-learning-to-detect-rare-objects","paper_url":"http://openaccess.thecvf.com/content_ICCV_2019/html/Wang_Meta-Learning_to_Detect_Rare_Objects_ICCV_2019_paper.html","paper_title":"Meta-Learning to Detect Rare Objects","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":32,"model":"MetaYOLO","metrics":{"AP":"5.6"},"uses_additional_data":false,"paper_date":"2018-12-05","paper":"/paper/few-shot-object-detection-via-feature","paper_url":"https://arxiv.org/abs/1812.01866v2","paper_title":"Few-shot Object Detection via Feature Reweighting","code":"https://github.com/bingykang/Fewshot_Detection","n_code_links":4,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":33,"model":"LSTD (YOLO)","metrics":{"AP":"3.2"},"uses_additional_data":false,"paper_date":"2018-03-05","paper":"/paper/lstd-a-low-shot-transfer-detector-for-object","paper_url":"http://arxiv.org/abs/1803.01529v1","paper_title":"LSTD: A Low-Shot Transfer Detector for Object Detection","code":"https://github.com/Cassie94/LSTD","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,795 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":15,"rows_with_any_sample_ran":14,"distinct_papers_with_graph_line":13,"distinct_papers_with_any_sample_ran":12,"samples_over_distinct_papers":{"n_ran":53,"n_unverified":52,"n_samples":105,"n_pointer_only_licence":20,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":59,"n_unverified":60,"n_samples":119,"n_pointer_only_licence":31,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}