{"url":"/sota/object-detection-on-coco-2017-val","task":{"name":"Object Detection","url":"/task/object-detection","note":null},"dataset":{"name":"COCO 2017 val","url":null},"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","AP50","AP75","APL","APM","APS","Param."],"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","AP50":null,"AP75":null,"APL":null,"APM":null,"APS":null,"Param.":null}},"counts":{"rows":33,"rows_with_code":33,"rows_with_paper_page":33,"rows_dated":33,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"Mr. DETR (Swin-L, 1x, 5cale)","metrics":{"AP":"61.8","AP50":"79.0","AP75":"67.6","APL":"75.7","APM":"65.6","APS":"47.7"},"uses_additional_data":true,"paper_date":"2024-12-13","paper":"/paper/mr-detr-instructive-multi-route-training-for","paper_url":"https://arxiv.org/abs/2412.10028v2","paper_title":"Mr. DETR: Instructive Multi-Route Training for Detection Transformers","code":"https://github.com/Visual-AI/Mr.DETR","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"Mr. DETR (Swin-L, 1x, 4scale)","metrics":{"AP":"58.4","AP50":"76.3","AP75":"63.9","APL":"75.3","APM":"62.8","APS":"40.8"},"uses_additional_data":false,"paper_date":"2024-12-13","paper":"/paper/mr-detr-instructive-multi-route-training-for","paper_url":"https://arxiv.org/abs/2412.10028v2","paper_title":"Mr. DETR: Instructive Multi-Route Training for Detection Transformers","code":"https://github.com/Visual-AI/Mr.DETR","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"MI-DETR (Swin-L 1x)","metrics":{"AP":"58.2","AP50":"76.5","AP75":"63.4","APL":"74.6","APM":"62.8","APS":"42.5"},"uses_additional_data":false,"paper_date":"2025-03-03","paper":"/paper/mi-detr-an-object-detection-model-with-multi","paper_url":"https://arxiv.org/abs/2503.01463v1","paper_title":"MI-DETR: An Object Detection Model with Multi-time Inquiries Mechanism","code":"https://github.com/CQU-ADHRI-Lab/MI-DETR","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"Relation-DETR (Swin-L 2x)","metrics":{"AP":"58.1","AP50":"76.4","AP75":"63.5","APL":"73.5","APM":"63.0","APS":"41.8"},"uses_additional_data":false,"paper_date":"2024-07-16","paper":"/paper/relation-detr-exploring-explicit-position","paper_url":"https://arxiv.org/abs/2407.11699v1","paper_title":"Relation DETR: Exploring Explicit Position Relation Prior for Object Detection","code":"https://github.com/xiuqhou/relation-detr","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"Relation-DETR (Swin-L 1x)","metrics":{"AP":"57.8","AP50":"76.1","AP75":"62.9","APL":"74.4","APM":"62.1","APS":"41.2"},"uses_additional_data":false,"paper_date":"2024-07-16","paper":"/paper/relation-detr-exploring-explicit-position","paper_url":"https://arxiv.org/abs/2407.11699v1","paper_title":"Relation DETR: Exploring Explicit Position Relation Prior for Object Detection","code":"https://github.com/xiuqhou/relation-detr","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"Salience-DETR (Focal-L 1x)","metrics":{"AP":"57.3","AP50":"75.5","AP75":"62.3","APL":"74.5","APM":"61.8","APS":"40.9","Param.":"220M"},"uses_additional_data":false,"paper_date":"2024-03-24","paper":"/paper/salience-detr-enhancing-detection-transformer-1","paper_url":"https://arxiv.org/abs/2403.16131v1","paper_title":"Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement","code":"https://github.com/xiuqhou/relation-detr","n_code_links":3,"syntology":{"n_ran":11,"n_unverified":12,"n_samples":23,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"YOLOv6-L6(46 fps, V100, bs1)","metrics":{"AP":"57.2","AP50":"74.5"},"uses_additional_data":false,"paper_date":"2023-01-13","paper":"/paper/yolov6-v3-0-a-full-scale-reloading","paper_url":"https://arxiv.org/abs/2301.05586v1","paper_title":"YOLOv6 v3.0: A Full-Scale Reloading","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":5,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":8,"model":"Salience-DETR (Swin-L 1x)","metrics":{"AP":"56.5","AP50":"75.0","AP75":"61.5","APL":"72.8","APM":"61.2","APS":"40.2","Param.":"210M"},"uses_additional_data":false,"paper_date":"2024-03-24","paper":"/paper/salience-detr-enhancing-detection-transformer-1","paper_url":"https://arxiv.org/abs/2403.16131v1","paper_title":"Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement","code":"https://github.com/xiuqhou/relation-detr","n_code_links":3,"syntology":{"n_ran":11,"n_unverified":12,"n_samples":23,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"MogaNet-XL (Cascade Mask R-CNN)","metrics":{"AP":"56.2"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"MogaNet-L (Cascade Mask R-CNN)","metrics":{"AP":"53.3"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"MogaNet-B (Cascade Mask R-CNN)","metrics":{"AP":"52.6"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"Relation-DETR (ResNet50 2x)","metrics":{"AP":"52.1","AP50":"69.7","AP75":"56.6","APL":"66.5","APM":"56.0","APS":"36.1"},"uses_additional_data":false,"paper_date":"2024-07-16","paper":"/paper/relation-detr-exploring-explicit-position","paper_url":"https://arxiv.org/abs/2407.11699v1","paper_title":"Relation DETR: Exploring Explicit Position Relation Prior for Object Detection","code":"https://github.com/xiuqhou/relation-detr","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"Relation-DETR (ResNet50 1x)","metrics":{"AP":"51.7","AP50":"69.1","AP75":"56.3","APL":"66.1","APM":"55.6","APS":"36.1"},"uses_additional_data":false,"paper_date":"2024-07-16","paper":"/paper/relation-detr-exploring-explicit-position","paper_url":"https://arxiv.org/abs/2407.11699v1","paper_title":"Relation DETR: Exploring Explicit Position Relation Prior for Object Detection","code":"https://github.com/xiuqhou/relation-detr","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"MogaNet-S (Cascade Mask R-CNN)","metrics":{"AP":"51.6"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"RF-ConvNeXt-T Cascade R-CNN","metrics":{"AP":"50.9"},"uses_additional_data":false,"paper_date":"2022-06-14","paper":"/paper/rf-next-efficient-receptive-field-search-for","paper_url":"https://arxiv.org/abs/2206.06637v2","paper_title":"RF-Next: Efficient Receptive Field Search for Convolutional Neural Networks","code":"https://github.com/ShangHua-Gao/G2L-search","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":16,"model":"Salience-DETR (ResNet50 1x)","metrics":{"AP":"50.0","AP50":"67.7","AP75":"54.2","APL":"64.4","APM":"54.4","APS":"33.3","Param.":"56M"},"uses_additional_data":false,"paper_date":"2024-03-24","paper":"/paper/salience-detr-enhancing-detection-transformer-1","paper_url":"https://arxiv.org/abs/2403.16131v1","paper_title":"Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement","code":"https://github.com/xiuqhou/relation-detr","n_code_links":3,"syntology":{"n_ran":11,"n_unverified":12,"n_samples":23,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"SQR-Adamixer-R101","metrics":{"AP":"49.8"},"uses_additional_data":false,"paper_date":"2022-12-15","paper":"/paper/enhanced-training-of-query-based-object","paper_url":"https://arxiv.org/abs/2212.07593v3","paper_title":"Enhanced Training of Query-Based Object Detection via Selective Query Recollection","code":"https://github.com/IDEA-Research/detrex","n_code_links":2,"syntology":null},{"rank_in_archive_order":18,"model":"MogaNet-L (Mask R-CNN 1x)","metrics":{"AP":"49.4"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"ViDT Swin-base","metrics":{"AP":"49.2","AP50":"69.4","AP75":"53.1","APL":"66.9","APM":"52.6","APS":"30.6","Param.":"0.1B"},"uses_additional_data":false,"paper_date":"2021-10-08","paper":"/paper/vidt-an-efficient-and-effective-fully","paper_url":"https://arxiv.org/abs/2110.03921v2","paper_title":"ViDT: An Efficient and Effective Fully Transformer-based Object Detector","code":"https://github.com/naver-ai/vidt","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"SQR-Adamixer-R50","metrics":{"AP":"48.9"},"uses_additional_data":false,"paper_date":"2022-12-15","paper":"/paper/enhanced-training-of-query-based-object","paper_url":"https://arxiv.org/abs/2212.07593v3","paper_title":"Enhanced Training of Query-Based Object Detection via Selective Query Recollection","code":"https://github.com/IDEA-Research/detrex","n_code_links":2,"syntology":null},{"rank_in_archive_order":21,"model":"MogaNet-L (RetinaNet 1x)","metrics":{"AP":"48.7"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"MogaNet-B (Mask R-CNN 1x)","metrics":{"AP":"47.9"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"MogaNet-B (RetinaNet 1x)","metrics":{"AP":"47.7"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":24,"model":"ViDT Swin-small","metrics":{"AP":"47.5","AP50":"67.7","AP75":"51.4","APL":"64.8","APM":"50.7","APS":"29.2","Param.":"61M"},"uses_additional_data":false,"paper_date":"2021-10-08","paper":"/paper/vidt-an-efficient-and-effective-fully","paper_url":"https://arxiv.org/abs/2110.03921v2","paper_title":"ViDT: An Efficient and Effective Fully Transformer-based Object Detector","code":"https://github.com/naver-ai/vidt","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"MogaNet-S (Mask R-CNN 1x)","metrics":{"AP":"46.7"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"MogaNet-S (RetinaNet 1x)","metrics":{"AP":"45.8"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"ViDT Swin-tiny","metrics":{"AP":"44.8","AP50":"64.5","AP75":"48.7","APL":"62.1","APM":"47.6","APS":"25.9","Param.":"38M"},"uses_additional_data":false,"paper_date":"2021-10-08","paper":"/paper/vidt-an-efficient-and-effective-fully","paper_url":"https://arxiv.org/abs/2110.03921v2","paper_title":"ViDT: An Efficient and Effective Fully Transformer-based Object Detector","code":"https://github.com/naver-ai/vidt","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"MogaNet-T (Mask R-CNN 1x)","metrics":{"AP":"42.6"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":29,"model":"MogaNet-T (RetinaNet 1x)","metrics":{"AP":"41.4"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":30,"model":"MogaNet-XT (Mask R-CNN 1x)","metrics":{"AP":"40.7"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":31,"model":"ViDT Swin-nano","metrics":{"AP":"40.4","AP50":"59.6","AP75":"43.3","APL":"55.8","APM":"42.5","APS":"23.2","Param.":"16M"},"uses_additional_data":false,"paper_date":"2021-10-08","paper":"/paper/vidt-an-efficient-and-effective-fully","paper_url":"https://arxiv.org/abs/2110.03921v2","paper_title":"ViDT: An Efficient and Effective Fully Transformer-based Object Detector","code":"https://github.com/naver-ai/vidt","n_code_links":1,"syntology":null},{"rank_in_archive_order":32,"model":"MogaNet-XT (RetinaNet 1x)","metrics":{"AP":"39.7"},"uses_additional_data":false,"paper_date":"2022-11-07","paper":"/paper/efficient-multi-order-gated-aggregation","paper_url":"https://arxiv.org/abs/2211.03295v3","paper_title":"MogaNet: Multi-order Gated Aggregation Network","code":"https://github.com/chengtan9907/OpenSTL","n_code_links":7,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":33,"model":"DyHead (Swin-T, multi scale)","metrics":{"AP50":"68","AP75":"54.3","APL":"64.2"},"uses_additional_data":false,"paper_date":"2021-06-15","paper":"/paper/dynamic-head-unifying-object-detection-heads","paper_url":"https://arxiv.org/abs/2106.08322v1","paper_title":"Dynamic Head: Unifying Object Detection Heads with Attentions","code":"https://github.com/open-mmlab/mmdetection","n_code_links":3,"syntology":{"n_ran":7,"n_unverified":2,"n_samples":9,"n_pointer_only_licence":0}}],"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":24,"rows_with_any_sample_ran":24,"distinct_papers_with_graph_line":6,"distinct_papers_with_any_sample_ran":6,"samples_over_distinct_papers":{"n_ran":35,"n_unverified":18,"n_samples":53,"n_pointer_only_licence":5,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":216,"n_unverified":81,"n_samples":297,"n_pointer_only_licence":5,"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"}}}