{"url":"/sota/object-detection-on-pascal-voc-2007","task":{"name":"Object Detection","url":"/task/object-detection","note":null},"dataset":{"name":"PASCAL VOC 2007","url":"/dataset/pascal-voc-2007"},"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":["MAP","AP50","mAP@50","mAP@50-95","box AP"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"MAP":"higher","AP50":null,"mAP@50":"higher","mAP@50-95":"higher","box AP":"higher"}},"counts":{"rows":30,"rows_with_code":29,"rows_with_paper_page":30,"rows_dated":30,"rows_using_additional_data":9},"rows":[{"rank_in_archive_order":1,"model":"Cascade Eff-B7 NAS-FPN (Copy Paste pre-training, single-scale)","metrics":{"MAP":"89.3%"},"uses_additional_data":false,"paper_date":"2020-12-13","paper":"/paper/simple-copy-paste-is-a-strong-data","paper_url":"https://arxiv.org/abs/2012.07177v2","paper_title":"Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation","code":"https://github.com/PaddlePaddle/PaddleOCR","n_code_links":5,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"YOLO-Former","metrics":{"MAP":"86.01%"},"uses_additional_data":true,"paper_date":"2024-01-11","paper":"/paper/yolo-former-yolo-shakes-hand-with-vit","paper_url":"https://arxiv.org/abs/2401.06244v1","paper_title":"YOLO-Former: YOLO Shakes Hand With ViT","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"DETReg (MDef-DETR)","metrics":{"AP50":"84.16","MAP":"84.16%"},"uses_additional_data":true,"paper_date":"2021-11-22","paper":"/paper/multi-modal-transformers-excel-at-class","paper_url":"https://arxiv.org/abs/2111.11430v6","paper_title":"Class-agnostic Object Detection with Multi-modal Transformer","code":"https://github.com/mmaaz60/mvits_for_class_agnostic_od","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"HSD (VGG16, 512x512, single-scale test)","metrics":{"MAP":"83.0%"},"uses_additional_data":false,"paper_date":"2019-10-01","paper":"/paper/hierarchical-shot-detector","paper_url":"http://openaccess.thecvf.com/content_ICCV_2019/html/Cao_Hierarchical_Shot_Detector_ICCV_2019_paper.html","paper_title":"Hierarchical Shot Detector","code":"https://github.com/JialeCao001/HSD","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"CoupleNet","metrics":{"MAP":"82.7%"},"uses_additional_data":false,"paper_date":"2017-08-09","paper":"/paper/couplenet-coupling-global-structure-with","paper_url":"http://arxiv.org/abs/1708.02863v1","paper_title":"CoupleNet: Coupling Global Structure with Local Parts for Object Detection","code":"https://github.com/princewang1994/R-FCN.pytorch","n_code_links":3,"syntology":null},{"rank_in_archive_order":6,"model":"EEEA-Net-C2 (YOLOv4)","metrics":{"MAP":"81.8%"},"uses_additional_data":true,"paper_date":"2021-08-13","paper":"/paper/eeea-net-an-early-exit-evolutionary-neural","paper_url":"https://arxiv.org/abs/2108.06156v1","paper_title":"EEEA-Net: An Early Exit Evolutionary Neural Architecture Search","code":"https://github.com/chakkritte/eeea-net","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"HSD (VGG16, 320x320, single-scale test)","metrics":{"MAP":"81.7%"},"uses_additional_data":false,"paper_date":"2019-10-01","paper":"/paper/hierarchical-shot-detector","paper_url":"http://openaccess.thecvf.com/content_ICCV_2019/html/Cao_Hierarchical_Shot_Detector_ICCV_2019_paper.html","paper_title":"Hierarchical Shot Detector","code":"https://github.com/JialeCao001/HSD","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"SSD512 (07+12+COCO)","metrics":{"MAP":"81.6%"},"uses_additional_data":true,"paper_date":"2015-12-08","paper":"/paper/ssd-single-shot-multibox-detector","paper_url":"http://arxiv.org/abs/1512.02325v5","paper_title":"SSD: Single Shot MultiBox Detector","code":"https://github.com/open-mmlab/mmdetection","n_code_links":221,"syntology":{"n_ran":19,"n_unverified":112,"n_samples":131,"n_pointer_only_licence":5}},{"rank_in_archive_order":9,"model":"BlitzNet512 + seg (s8)","metrics":{"MAP":"81.5%"},"uses_additional_data":false,"paper_date":"2017-08-09","paper":"/paper/blitznet-a-real-time-deep-network-for-scene","paper_url":"http://arxiv.org/abs/1708.02813v1","paper_title":"BlitzNet: A Real-Time Deep Network for Scene Understanding","code":"https://github.com/dvornikita/blitznet","n_code_links":2,"syntology":null},{"rank_in_archive_order":10,"model":"Localize","metrics":{"MAP":"81.5%"},"uses_additional_data":false,"paper_date":"2020-09-29","paper":"/paper/localize-to-classify-and-classify-to-localize","paper_url":"https://arxiv.org/abs/2009.14085v1","paper_title":"Localize to Classify and Classify to Localize: Mutual Guidance in Object Detection","code":"https://github.com/ZHANGHeng19931123/MutualGuide","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"CenterNet(DLA34, Flip, 512x512)","metrics":{"MAP":"80.7%"},"uses_additional_data":false,"paper_date":"2019-04-16","paper":"/paper/objects-as-points","paper_url":"http://arxiv.org/abs/1904.07850v2","paper_title":"Objects as Points","code":"https://github.com/tensorflow/models/tree/master/research/object_detection","n_code_links":76,"syntology":{"n_ran":10,"n_unverified":120,"n_samples":130,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"PS-KD (ResNet-152, CutMix)","metrics":{"MAP":"79.7%"},"uses_additional_data":false,"paper_date":"2020-06-22","paper":"/paper/self-knowledge-distillation-a-simple-way-for","paper_url":"https://arxiv.org/abs/2006.12000v3","paper_title":"Self-Knowledge Distillation with Progressive Refinement of Targets","code":"https://github.com/lgcnsai/ps-kd-pytorch","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"DPNet","metrics":{"MAP":"79.2%"},"uses_additional_data":false,"paper_date":"2022-09-28","paper":"/paper/dpnet-dual-path-network-for-real-time-object","paper_url":"https://arxiv.org/abs/2209.13933v1","paper_title":"DPNet: Dual-Path Network for Real-time Object Detection with Lightweight Attention","code":"https://github.com/huiminshii/dpnet","n_code_links":2,"syntology":null},{"rank_in_archive_order":14,"model":"OHEM","metrics":{"MAP":"78.9%"},"uses_additional_data":false,"paper_date":"2016-04-12","paper":"/paper/training-region-based-object-detectors-with","paper_url":"http://arxiv.org/abs/1604.03540v1","paper_title":"Training Region-based Object Detectors with Online Hard Example Mining","code":"https://github.com/abhi2610/ohem","n_code_links":5,"syntology":null},{"rank_in_archive_order":15,"model":"YOLO v2","metrics":{"MAP":"78.6%"},"uses_additional_data":true,"paper_date":"2016-12-25","paper":"/paper/yolo9000-better-faster-stronger","paper_url":"http://arxiv.org/abs/1612.08242v1","paper_title":"YOLO9000: Better, Faster, Stronger","code":"https://github.com/AlexeyAB/darknet","n_code_links":231,"syntology":{"n_ran":16,"n_unverified":44,"n_samples":60,"n_pointer_only_licence":22}},{"rank_in_archive_order":16,"model":"ThunderNet SNet535 Backbone","metrics":{"MAP":"78.6%"},"uses_additional_data":false,"paper_date":"2019-03-28","paper":"/paper/thundernet-towards-real-time-generic-object","paper_url":"https://arxiv.org/abs/1903.11752v3","paper_title":"ThunderNet: Towards Real-time Generic Object Detection","code":"https://github.com/ouyanghuiyu/Thundernet_Pytorch","n_code_links":3,"syntology":null},{"rank_in_archive_order":17,"model":"DeNet-101 (skip)","metrics":{"MAP":"77.1%"},"uses_additional_data":false,"paper_date":"2017-03-30","paper":"/paper/denet-scalable-real-time-object-detection","paper_url":"http://arxiv.org/abs/1703.10295v3","paper_title":"DeNet: Scalable Real-time Object Detection with Directed Sparse Sampling","code":"https://github.com/lachlants/denet","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"I+ORE","metrics":{"MAP":"76.2%"},"uses_additional_data":false,"paper_date":"2017-08-16","paper":"/paper/random-erasing-data-augmentation","paper_url":"http://arxiv.org/abs/1708.04896v2","paper_title":"Random Erasing Data Augmentation","code":"https://github.com/rwightman/pytorch-image-models","n_code_links":18,"syntology":null},{"rank_in_archive_order":19,"model":"Perona Malik (Perona and Malik, 1990)","metrics":{"MAP":"74.37%"},"uses_additional_data":false,"paper_date":"2020-11-03","paper":"/paper/learning-visual-representations-for-transfer-1","paper_url":"https://arxiv.org/abs/2011.01901v3","paper_title":"Learning Visual Representations for Transfer Learning by Suppressing Texture","code":"https://github.com/HaohanWang/ImageNet-Sketch","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"FRCN","metrics":{"MAP":"74.2%"},"uses_additional_data":false,"paper_date":"2017-04-11","paper":"/paper/a-fast-rcnn-hard-positive-generation-via","paper_url":"http://arxiv.org/abs/1704.03414v1","paper_title":"A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection","code":"https://github.com/xiaolonw/adversarial-frcnn","n_code_links":4,"syntology":null},{"rank_in_archive_order":21,"model":"VGG-16 + KL Loss + var voting + soft-NMS","metrics":{"MAP":"71.6%"},"uses_additional_data":false,"paper_date":"2018-09-23","paper":"/paper/softer-nms-rethinking-bounding-box-regression","paper_url":"http://arxiv.org/abs/1809.08545v3","paper_title":"Bounding Box Regression with Uncertainty for Accurate Object Detection","code":"https://github.com/yihui-he/KL-Loss","n_code_links":4,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"Fast R-CNN","metrics":{"MAP":"70.0%"},"uses_additional_data":true,"paper_date":"2015-04-30","paper":"/paper/fast-r-cnn","paper_url":"http://arxiv.org/abs/1504.08083v2","paper_title":"Fast R-CNN","code":"https://github.com/facebookresearch/detectron","n_code_links":30,"syntology":{"n_ran":5,"n_unverified":3,"n_samples":8,"n_pointer_only_licence":1}},{"rank_in_archive_order":23,"model":"subCNN","metrics":{"MAP":"68.5%"},"uses_additional_data":false,"paper_date":"2016-04-16","paper":"/paper/subcategory-aware-convolutional-neural","paper_url":"http://arxiv.org/abs/1604.04693v3","paper_title":"Subcategory-aware Convolutional Neural Networks for Object Proposals and Detection","code":"https://github.com/xiaohaoChen/rrc_detection","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"YOLO","metrics":{"MAP":"63.4%"},"uses_additional_data":true,"paper_date":"2015-06-08","paper":"/paper/you-only-look-once-unified-real-time-object","paper_url":"http://arxiv.org/abs/1506.02640v5","paper_title":"You Only Look Once: Unified, Real-Time Object Detection","code":"https://github.com/AlexeyAB/darknet","n_code_links":144,"syntology":{"n_ran":80,"n_unverified":68,"n_samples":148,"n_pointer_only_licence":98}},{"rank_in_archive_order":25,"model":"SPP(combination)","metrics":{"MAP":"60.9%"},"uses_additional_data":false,"paper_date":"2014-06-18","paper":"/paper/spatial-pyramid-pooling-in-deep-convolutional","paper_url":"http://arxiv.org/abs/1406.4729v4","paper_title":"Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition","code":"https://github.com/yhenon/keras-spp","n_code_links":14,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"R-CNN","metrics":{"MAP":"58.5%"},"uses_additional_data":true,"paper_date":"2013-11-11","paper":"/paper/rich-feature-hierarchies-for-accurate-object","paper_url":"http://arxiv.org/abs/1311.2524v5","paper_title":"Rich feature hierarchies for accurate object detection and semantic segmentation","code":"https://github.com/rbgirshick/rcnn","n_code_links":29,"syntology":{"n_ran":1,"n_unverified":16,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"Deformable Parts Model (DeepPyramid)","metrics":{"MAP":"45.2%"},"uses_additional_data":false,"paper_date":"2014-09-18","paper":"/paper/deformable-part-models-are-convolutional","paper_url":"http://arxiv.org/abs/1409.5403v2","paper_title":"Deformable Part Models are Convolutional Neural Networks","code":"https://github.com/rbgirshick/DeepPyramid","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"TinyissimoYOLO-v8","metrics":{"MAP":"42.3%"},"uses_additional_data":false,"paper_date":"2023-11-02","paper":"/paper/ultra-efficient-on-device-object-detection-on","paper_url":"https://arxiv.org/abs/2311.01057v2","paper_title":"Ultra-Efficient On-Device Object Detection on AI-Integrated Smart Glasses with TinyissimoYOLO","code":"https://github.com/eth-pbl/tinyissimoyolo","n_code_links":1,"syntology":null},{"rank_in_archive_order":29,"model":"FemotoDet","metrics":{"AP50":"46.31","MAP":"22.90%"},"uses_additional_data":false,"paper_date":"2023-01-17","paper":"/paper/femtodet-an-object-detection-baseline-for","paper_url":"https://arxiv.org/abs/2301.06719v5","paper_title":"FemtoDet: An Object Detection Baseline for Energy Versus Performance Tradeoffs","code":"https://github.com/yh-pengtu/FemtoDet","n_code_links":1,"syntology":null},{"rank_in_archive_order":30,"model":"YOLOv7+Inner-IoU","metrics":{"mAP@50":"64.44","mAP@50-95":"38.52"},"uses_additional_data":true,"paper_date":"2023-11-06","paper":"/paper/inner-iou-more-effective-intersection-over","paper_url":"https://arxiv.org/abs/2311.02877v4","paper_title":"Inner-IoU: More Effective Intersection over Union Loss with Auxiliary Bounding Box","code":"https://github.com/malagoutou/Inner-IoU","n_code_links":1,"syntology":null}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"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":10,"rows_with_any_sample_ran":8,"distinct_papers_with_graph_line":10,"distinct_papers_with_any_sample_ran":8,"samples_over_distinct_papers":{"n_ran":137,"n_unverified":369,"n_samples":506,"n_pointer_only_licence":126,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":137,"n_unverified":369,"n_samples":506,"n_pointer_only_licence":126,"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"}}}