{"url":"/sota/object-detection-on-coco-2017","task":{"name":"Object Detection","url":"/task/object-detection","note":null},"dataset":{"name":"COCO 2017","url":"/dataset/coco"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":null,"description_from":null,"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","mAP","Mean mAP","AP50","AP75","APM","APM50","APM75"],"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","mAP":"higher","Mean mAP":"higher","AP50":null,"AP75":null,"APM":null,"APM50":null,"APM75":null}},"counts":{"rows":24,"rows_with_code":21,"rows_with_paper_page":24,"rows_dated":24,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"MaxViT-B","metrics":{"AP":"53.4","AP50":"72.9","AP75":"58.1","APM":"45.7","APM50":"70.3","APM75":"50"},"uses_additional_data":false,"paper_date":"2022-04-04","paper":"/paper/maxvit-multi-axis-vision-transformer","paper_url":"https://arxiv.org/abs/2204.01697v4","paper_title":"MaxViT: Multi-Axis Vision Transformer","code":"https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/maxxvit.py","n_code_links":15,"syntology":{"n_ran":33,"n_unverified":20,"n_samples":53,"n_pointer_only_licence":9}},{"rank_in_archive_order":2,"model":"MaxViT-S","metrics":{"AP":"53.1","AP50":"72.5","AP75":"58.1","APM":"45.4","APM50":"69.8","APM75":"49.5"},"uses_additional_data":false,"paper_date":"2022-04-04","paper":"/paper/maxvit-multi-axis-vision-transformer","paper_url":"https://arxiv.org/abs/2204.01697v4","paper_title":"MaxViT: Multi-Axis Vision Transformer","code":"https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/maxxvit.py","n_code_links":15,"syntology":{"n_ran":33,"n_unverified":20,"n_samples":53,"n_pointer_only_licence":9}},{"rank_in_archive_order":3,"model":"MaxViT-T","metrics":{"AP":"52.1","AP50":"71.9","AP75":"56.8","APM":"44.6","APM50":"69.1","APM75":"48.4"},"uses_additional_data":false,"paper_date":"2022-04-04","paper":"/paper/maxvit-multi-axis-vision-transformer","paper_url":"https://arxiv.org/abs/2204.01697v4","paper_title":"MaxViT: Multi-Axis Vision Transformer","code":"https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/maxxvit.py","n_code_links":15,"syntology":{"n_ran":33,"n_unverified":20,"n_samples":53,"n_pointer_only_licence":9}},{"rank_in_archive_order":4,"model":"DAT-S++","metrics":{"AP":"50.2"},"uses_additional_data":false,"paper_date":"2023-09-04","paper":"/paper/dat-spatially-dynamic-vision-transformer-with","paper_url":"https://arxiv.org/abs/2309.01430v1","paper_title":"DAT++: Spatially Dynamic Vision Transformer with Deformable Attention","code":"https://github.com/leaplabthu/dat","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":8,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"DAT-T++","metrics":{"AP":"49.2"},"uses_additional_data":false,"paper_date":"2023-09-04","paper":"/paper/dat-spatially-dynamic-vision-transformer-with","paper_url":"https://arxiv.org/abs/2309.01430v1","paper_title":"DAT++: Spatially Dynamic Vision Transformer with Deformable Attention","code":"https://github.com/leaplabthu/dat","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":8,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"DyHead (SAP)","metrics":{"AP":"42.1","AP50":"59.4","AP75":"45.9"},"uses_additional_data":false,"paper_date":"2024-09-25","paper":"/paper/stochastic-subsampling-with-average-pooling","paper_url":"https://arxiv.org/abs/2409.16630v1","paper_title":"Stochastic Subsampling With Average Pooling","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"Faster R-CNN (ideal number of groups)","metrics":{"AP":"40.7","AP50":"61.2","AP75":"44.6"},"uses_additional_data":false,"paper_date":"2023-02-07","paper":"/paper/on-the-ideal-number-of-groups-for-isometric","paper_url":"https://arxiv.org/abs/2302.03193v1","paper_title":"On the Ideal Number of Groups for Isometric Gradient Propagation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"UniRepLKNet-XL++","metrics":{"mAP":"56.4"},"uses_additional_data":false,"paper_date":"2023-11-27","paper":"/paper/unireplknet-a-universal-perception-large","paper_url":"https://arxiv.org/abs/2311.15599v2","paper_title":"UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition","code":"https://github.com/ailab-cvc/unireplknet","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"UniRepLKNet-L++","metrics":{"mAP":"55.8"},"uses_additional_data":false,"paper_date":"2023-11-27","paper":"/paper/unireplknet-a-universal-perception-large","paper_url":"https://arxiv.org/abs/2311.15599v2","paper_title":"UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition","code":"https://github.com/ailab-cvc/unireplknet","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"UniRepLKNet-B++","metrics":{"mAP":"54.8"},"uses_additional_data":false,"paper_date":"2023-11-27","paper":"/paper/unireplknet-a-universal-perception-large","paper_url":"https://arxiv.org/abs/2311.15599v2","paper_title":"UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition","code":"https://github.com/ailab-cvc/unireplknet","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"UniRepLKNet-S++","metrics":{"mAP":"54.3"},"uses_additional_data":false,"paper_date":"2023-11-27","paper":"/paper/unireplknet-a-universal-perception-large","paper_url":"https://arxiv.org/abs/2311.15599v2","paper_title":"UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition","code":"https://github.com/ailab-cvc/unireplknet","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"MixMIM-L","metrics":{"mAP":"54.1"},"uses_additional_data":false,"paper_date":"2022-05-26","paper":"/paper/mixmim-mixed-and-masked-image-modeling-for","paper_url":"https://arxiv.org/abs/2205.13137v4","paper_title":"MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers","code":"https://github.com/sense-x/mixmim","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"UniRepLKNet-S","metrics":{"mAP":"53"},"uses_additional_data":false,"paper_date":"2023-11-27","paper":"/paper/unireplknet-a-universal-perception-large","paper_url":"https://arxiv.org/abs/2311.15599v2","paper_title":"UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition","code":"https://github.com/ailab-cvc/unireplknet","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"MixMIM-B","metrics":{"mAP":"52.2"},"uses_additional_data":false,"paper_date":"2022-05-26","paper":"/paper/mixmim-mixed-and-masked-image-modeling-for","paper_url":"https://arxiv.org/abs/2205.13137v4","paper_title":"MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers","code":"https://github.com/sense-x/mixmim","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"UniRepLKNet-T","metrics":{"mAP":"51.7"},"uses_additional_data":false,"paper_date":"2023-11-27","paper":"/paper/unireplknet-a-universal-perception-large","paper_url":"https://arxiv.org/abs/2311.15599v2","paper_title":"UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition","code":"https://github.com/ailab-cvc/unireplknet","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"BiFormer-B (IN1k pretrain, MaskRCNN 12ep)","metrics":{"mAP":"48.6"},"uses_additional_data":false,"paper_date":"2023-03-15","paper":"/paper/biformer-vision-transformer-with-bi-level","paper_url":"https://arxiv.org/abs/2303.08810v1","paper_title":"BiFormer: Vision Transformer with Bi-Level Routing Attention","code":"https://github.com/rayleizhu/biformer","n_code_links":3,"syntology":{"n_ran":6,"n_unverified":0,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"DeBiFormer-B (IN1k pretrain, MaskRCNN 12ep)","metrics":{"mAP":"48.5"},"uses_additional_data":false,"paper_date":"2024-10-11","paper":"/paper/debiformer-vision-transformer-with-deformable","paper_url":"https://arxiv.org/abs/2410.08582v1","paper_title":"DeBiFormer: Vision Transformer with Deformable Agent Bi-level Routing Attention","code":"https://github.com/maclong01/DeBiFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"BiFormer-S (IN1k pretrain, MaskRCNN 12ep)","metrics":{"mAP":"47.8"},"uses_additional_data":false,"paper_date":"2023-03-15","paper":"/paper/biformer-vision-transformer-with-bi-level","paper_url":"https://arxiv.org/abs/2303.08810v1","paper_title":"BiFormer: Vision Transformer with Bi-Level Routing Attention","code":"https://github.com/rayleizhu/biformer","n_code_links":3,"syntology":{"n_ran":6,"n_unverified":0,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"DeBiFormer-S (IN1k pretrain, MaskRCNN 12ep)","metrics":{"mAP":"47.5"},"uses_additional_data":false,"paper_date":"2024-10-11","paper":"/paper/debiformer-vision-transformer-with-deformable","paper_url":"https://arxiv.org/abs/2410.08582v1","paper_title":"DeBiFormer: Vision Transformer with Deformable Agent Bi-level Routing Attention","code":"https://github.com/maclong01/DeBiFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"DeBiFormer-B (IN1k pretrain, Retina)","metrics":{"mAP":"47.1"},"uses_additional_data":false,"paper_date":"2024-10-11","paper":"/paper/debiformer-vision-transformer-with-deformable","paper_url":"https://arxiv.org/abs/2410.08582v1","paper_title":"DeBiFormer: Vision Transformer with Deformable Agent Bi-level Routing Attention","code":"https://github.com/maclong01/DeBiFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"DeBiFormer-S (IN1k pretrain, Retina)","metrics":{"mAP":"45.6"},"uses_additional_data":false,"paper_date":"2024-10-11","paper":"/paper/debiformer-vision-transformer-with-deformable","paper_url":"https://arxiv.org/abs/2410.08582v1","paper_title":"DeBiFormer: Vision Transformer with Deformable Agent Bi-level Routing Attention","code":"https://github.com/maclong01/DeBiFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"YOLO-Drone","metrics":{"mAP":"35.45"},"uses_additional_data":false,"paper_date":"2023-04-14","paper":"/paper/yolo-drone-airborne-real-time-detection-of","paper_url":"https://arxiv.org/abs/2304.06925v2","paper_title":"YOLO-Drone:Airborne real-time detection of dense small objects from high-altitude perspective","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":23,"model":"retinanet","metrics":{"Mean mAP":"3153"},"uses_additional_data":false,"paper_date":"2019-12-19","paper":"/paper/benchmark-for-generic-product-detection-a","paper_url":"https://arxiv.org/abs/1912.09476v2","paper_title":"Benchmark for Generic Product Detection: A Low Data Baseline for Dense Object Detection","code":"https://github.com/ParallelDots/generic-sku-detection-benchmark","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"Lpixel","metrics":{"Mean mAP":"4.2"},"uses_additional_data":false,"paper_date":"2021-08-09","paper":"/paper/paint-transformer-feed-forward-neural","paper_url":"https://arxiv.org/abs/2108.03798v2","paper_title":"Paint Transformer: Feed Forward Neural Painting with Stroke Prediction","code":"https://github.com/huage001/painttransformer","n_code_links":2,"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 6,264 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":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"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":13,"rows_with_any_sample_ran":13,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":42,"n_unverified":29,"n_samples":71,"n_pointer_only_licence":9,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":121,"n_unverified":82,"n_samples":203,"n_pointer_only_licence":27,"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"}}}