{"url":"/sota/real-time-object-detection-on-coco","task":{"name":"Real-Time Object Detection","url":"/task/real-time-object-detection","note":null},"dataset":{"name":"COCO (Common Objects in Context)","url":"/dataset/coco"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**Real-Time Object Detection** is a computer vision task that involves identifying and locating objects of interest in real-time video sequences with fast inference while maintaining a base level of accuracy.\r\n\r\nThis is typically solved using algorithms that combine object detection and tracking techniques to accurately detect and track objects in real-time. They use a combination of feature extraction, object proposal generation, and classification to detect and localize objects of interest.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [CenterNet](https://github.com/xingyizhou/CenterNet) )</span>","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":["box AP","FPS (V100, b=1)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"box AP":"higher","FPS (V100, b=1)":null}},"counts":{"rows":82,"rows_with_code":76,"rows_with_paper_page":82,"rows_dated":82,"rows_using_additional_data":12},"rows":[{"rank_in_archive_order":1,"model":"DEIM-D-FINE-X+","metrics":{"FPS (V100, b=1)":"78 (T4)","box AP":"59.5"},"uses_additional_data":false,"paper_date":"2024-12-05","paper":"/paper/deim-detr-with-improved-matching-for-fast","paper_url":"https://arxiv.org/abs/2412.04234v3","paper_title":"DEIM: DETR with Improved Matching for Fast Convergence","code":"https://github.com/shihuahuang95/deim","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":12,"n_samples":13,"n_pointer_only_licence":13}},{"rank_in_archive_order":2,"model":"D-FINE-X+","metrics":{"FPS (V100, b=1)":"78 (T4)","box AP":"59.3"},"uses_additional_data":false,"paper_date":"2024-10-17","paper":"/paper/d-fine-redefine-regression-task-in-detrs-as","paper_url":"https://arxiv.org/abs/2410.13842v1","paper_title":"D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement","code":"https://github.com/huggingface/transformers","n_code_links":5,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"YOLOv6-L6(1280)","metrics":{"FPS (V100, b=1)":"26","box AP":"57.2"},"uses_additional_data":false,"paper_date":"2022-09-07","paper":"/paper/yolov6-a-single-stage-object-detection","paper_url":"https://arxiv.org/abs/2209.02976v1","paper_title":"YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":7,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":4,"model":"D-FINE-L+","metrics":{"FPS (V100, b=1)":"124 (T4)","box AP":"57.1"},"uses_additional_data":false,"paper_date":"2022-06-19","paper":"/paper/0-1-deep-neural-networks-via-block-coordinate","paper_url":"https://arxiv.org/abs/2206.09379v2","paper_title":"0/1 Deep Neural Networks via Block Coordinate Descent","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"PRB-FPN6-E-ELAN(1280)","metrics":{"FPS (V100, b=1)":"31","box AP":"56.9"},"uses_additional_data":false,"paper_date":"2020-12-03","paper":"/paper/parallel-residual-bi-fusion-feature-pyramid","paper_url":"https://arxiv.org/abs/2012.01724v5","paper_title":"Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate Single-Shot Object Detection","code":"https://github.com/pingyang1117/PRBNet_PyTorch","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"YOLOv7-E6E(1280)","metrics":{"FPS (V100, b=1)":"36","box AP":"56.8"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/yolov7-trainable-bag-of-freebies-sets-new","paper_url":"https://arxiv.org/abs/2207.02696v1","paper_title":"YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","code":"https://github.com/pjreddie/darknet","n_code_links":21,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"YOLOv7-D6(1280)","metrics":{"FPS (V100, b=1)":"44","box AP":"56.6"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/yolov7-trainable-bag-of-freebies-sets-new","paper_url":"https://arxiv.org/abs/2207.02696v1","paper_title":"YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","code":"https://github.com/pjreddie/darknet","n_code_links":21,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"DEIM-D-FINE-X","metrics":{"FPS (V100, b=1)":"78 (T4)","box AP":"56.5"},"uses_additional_data":false,"paper_date":"2024-12-05","paper":"/paper/deim-detr-with-improved-matching-for-fast","paper_url":"https://arxiv.org/abs/2412.04234v3","paper_title":"DEIM: DETR with Improved Matching for Fast Convergence","code":"https://github.com/shihuahuang95/deim","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":12,"n_samples":13,"n_pointer_only_licence":13}},{"rank_in_archive_order":9,"model":"RT-DETR-H(640)","metrics":{"FPS (V100, b=1)":"40 (T4)","box AP":"56.3"},"uses_additional_data":false,"paper_date":"2023-04-17","paper":"/paper/detrs-beat-yolos-on-real-time-object","paper_url":"https://arxiv.org/abs/2304.08069v3","paper_title":"DETRs Beat YOLOs on Real-time Object Detection","code":"https://github.com/huggingface/transformers","n_code_links":9,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"YOLOv7-E6(1280)","metrics":{"FPS (V100, b=1)":"56","box AP":"56"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/yolov7-trainable-bag-of-freebies-sets-new","paper_url":"https://arxiv.org/abs/2207.02696v1","paper_title":"YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","code":"https://github.com/pjreddie/darknet","n_code_links":21,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"D-FINE-X","metrics":{"FPS (V100, b=1)":"78 (T4)","box AP":"55.8"},"uses_additional_data":false,"paper_date":"2024-10-17","paper":"/paper/d-fine-redefine-regression-task-in-detrs-as","paper_url":"https://arxiv.org/abs/2410.13842v1","paper_title":"D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement","code":"https://github.com/huggingface/transformers","n_code_links":5,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"YOLOv9-E","metrics":{"box AP":"55.6"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":13,"model":"YOLOR-D6","metrics":{"FPS (V100, b=1)":"30","box AP":"55.4"},"uses_additional_data":true,"paper_date":"2021-05-10","paper":"/paper/you-only-learn-one-representation-unified","paper_url":"https://arxiv.org/abs/2105.04206v1","paper_title":"You Only Learn One Representation: Unified Network for Multiple Tasks","code":"https://github.com/WongKinYiu/yolor","n_code_links":9,"syntology":null},{"rank_in_archive_order":14,"model":"YOLOv12x","metrics":{"FPS (V100, b=1)":"85 (T4)","box AP":"55.2"},"uses_additional_data":false,"paper_date":"2025-02-20","paper":"/paper/yolov12-a-breakdown-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2502.14740v1","paper_title":"YOLOv12: A Breakdown of the Key Architectural Features","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"D-FINE-M+","metrics":{"FPS (V100, b=1)":"178 (T4)","box AP":"55.1"},"uses_additional_data":true,"paper_date":"2024-10-17","paper":"/paper/d-fine-redefine-regression-task-in-detrs-as","paper_url":"https://arxiv.org/abs/2410.13842v1","paper_title":"D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement","code":"https://github.com/huggingface/transformers","n_code_links":5,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"GELAN-E","metrics":{"box AP":"55.0"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":17,"model":"YOLOv7-W6(1280)","metrics":{"FPS (V100, b=1)":"84","box AP":"54.9"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/yolov7-trainable-bag-of-freebies-sets-new","paper_url":"https://arxiv.org/abs/2207.02696v1","paper_title":"YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","code":"https://github.com/pjreddie/darknet","n_code_links":21,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":18,"model":"RT-DETR-X","metrics":{"FPS (V100, b=1)":"74 (T4)","box AP":"54.8"},"uses_additional_data":false,"paper_date":"2023-04-17","paper":"/paper/detrs-beat-yolos-on-real-time-object","paper_url":"https://arxiv.org/abs/2304.08069v3","paper_title":"DETRs Beat YOLOs on Real-time Object Detection","code":"https://github.com/huggingface/transformers","n_code_links":9,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"YOLOR-E6","metrics":{"FPS (V100, b=1)":"37","box AP":"54.8"},"uses_additional_data":true,"paper_date":"2021-05-10","paper":"/paper/you-only-learn-one-representation-unified","paper_url":"https://arxiv.org/abs/2105.04206v1","paper_title":"You Only Learn One Representation: Unified Network for Multiple Tasks","code":"https://github.com/WongKinYiu/yolor","n_code_links":9,"syntology":null},{"rank_in_archive_order":20,"model":"DEIM-D-FINE-L","metrics":{"FPS (V100, b=1)":"124 (T4)","box AP":"54.7"},"uses_additional_data":false,"paper_date":"2024-12-05","paper":"/paper/deim-detr-with-improved-matching-for-fast","paper_url":"https://arxiv.org/abs/2412.04234v3","paper_title":"DEIM: DETR with Improved Matching for Fast Convergence","code":"https://github.com/shihuahuang95/deim","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":12,"n_samples":13,"n_pointer_only_licence":13}},{"rank_in_archive_order":21,"model":"YOLOv11x","metrics":{"FPS (V100, b=1)":"88 (T4)","box AP":"54.7"},"uses_additional_data":false,"paper_date":"2024-10-23","paper":"/paper/yolov11-an-overview-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2410.17725v1","paper_title":"YOLOv11: An Overview of the Key Architectural Enhancements","code":"https://github.com/ultralytics/ultralytics","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"PP-YOLOE+_X","metrics":{"FPS (V100, b=1)":"45","box AP":"54.7"},"uses_additional_data":false,"paper_date":"2022-03-30","paper":"/paper/pp-yoloe-an-evolved-version-of-yolo","paper_url":"https://arxiv.org/abs/2203.16250v3","paper_title":"PP-YOLOE: An evolved version of YOLO","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":8,"syntology":{"n_ran":5,"n_unverified":22,"n_samples":27,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"YOLOv10-X","metrics":{"box AP":"54.4"},"uses_additional_data":false,"paper_date":"2024-05-23","paper":"/paper/yolov10-real-time-end-to-end-object-detection","paper_url":"https://arxiv.org/abs/2405.14458v2","paper_title":"YOLOv10: Real-Time End-to-End Object Detection","code":"https://github.com/THU-MIG/yolov10","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":24,"model":"RT-DETR-R101","metrics":{"FPS (V100, b=1)":"74 (T4)","box AP":"54.3"},"uses_additional_data":false,"paper_date":"2023-04-17","paper":"/paper/detrs-beat-yolos-on-real-time-object","paper_url":"https://arxiv.org/abs/2304.08069v3","paper_title":"DETRs Beat YOLOs on Real-time Object Detection","code":"https://github.com/huggingface/transformers","n_code_links":9,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":25,"model":"YOLOR-W6","metrics":{"FPS (V100, b=1)":"47","box AP":"54.1"},"uses_additional_data":true,"paper_date":"2021-05-10","paper":"/paper/you-only-learn-one-representation-unified","paper_url":"https://arxiv.org/abs/2105.04206v1","paper_title":"You Only Learn One Representation: Unified Network for Multiple Tasks","code":"https://github.com/WongKinYiu/yolor","n_code_links":9,"syntology":null},{"rank_in_archive_order":26,"model":"D-FINE-L","metrics":{"FPS (V100, b=1)":"124 (T4)","box AP":"54.0"},"uses_additional_data":false,"paper_date":"2024-10-17","paper":"/paper/d-fine-redefine-regression-task-in-detrs-as","paper_url":"https://arxiv.org/abs/2410.13842v1","paper_title":"D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement","code":"https://github.com/huggingface/transformers","n_code_links":5,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"PP-YOLOE+_L(distillation)","metrics":{"FPS (V100, b=1)":"78","box AP":"54.0"},"uses_additional_data":false,"paper_date":"2022-03-30","paper":"/paper/pp-yoloe-an-evolved-version-of-yolo","paper_url":"https://arxiv.org/abs/2203.16250v3","paper_title":"PP-YOLOE: An evolved version of YOLO","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":8,"syntology":{"n_ran":5,"n_unverified":22,"n_samples":27,"n_pointer_only_licence":0}},{"rank_in_archive_order":28,"model":"YOLOv12l","metrics":{"FPS (V100, b=1)":"148 (T4)","box AP":"53.7"},"uses_additional_data":false,"paper_date":"2025-02-20","paper":"/paper/yolov12-a-breakdown-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2502.14740v1","paper_title":"YOLOv12: A Breakdown of the Key Architectural Features","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":29,"model":"YOLOv11l","metrics":{"FPS (V100, b=1)":"161 (T4)","box AP":"53.4"},"uses_additional_data":false,"paper_date":"2024-10-23","paper":"/paper/yolov11-an-overview-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2410.17725v1","paper_title":"YOLOv11: An Overview of the Key Architectural Enhancements","code":"https://github.com/ultralytics/ultralytics","n_code_links":1,"syntology":null},{"rank_in_archive_order":30,"model":"YOLOv10-L","metrics":{"box AP":"53.4"},"uses_additional_data":false,"paper_date":"2024-05-23","paper":"/paper/yolov10-real-time-end-to-end-object-detection","paper_url":"https://arxiv.org/abs/2405.14458v2","paper_title":"YOLOv10: Real-Time End-to-End Object Detection","code":"https://github.com/THU-MIG/yolov10","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":31,"model":"PRB-FPN-MSP","metrics":{"FPS (V100, b=1)":"94","box AP":"53.3"},"uses_additional_data":false,"paper_date":"2020-12-03","paper":"/paper/parallel-residual-bi-fusion-feature-pyramid","paper_url":"https://arxiv.org/abs/2012.01724v5","paper_title":"Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate Single-Shot Object Detection","code":"https://github.com/pingyang1117/PRBNet_PyTorch","n_code_links":1,"syntology":null},{"rank_in_archive_order":32,"model":"YOLOv7-X","metrics":{"FPS (V100, b=1)":"114","box AP":"53.1"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/yolov7-trainable-bag-of-freebies-sets-new","paper_url":"https://arxiv.org/abs/2207.02696v1","paper_title":"YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","code":"https://github.com/pjreddie/darknet","n_code_links":21,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":33,"model":"RT-DETR-L","metrics":{"FPS (V100, b=1)":"114 (T4)","box AP":"53.0"},"uses_additional_data":false,"paper_date":"2023-04-17","paper":"/paper/detrs-beat-yolos-on-real-time-object","paper_url":"https://arxiv.org/abs/2304.08069v3","paper_title":"DETRs Beat YOLOs on Real-time Object Detection","code":"https://github.com/huggingface/transformers","n_code_links":9,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":34,"model":"YOLOR-P6D","metrics":{"FPS (V100, b=1)":"49","box AP":"53"},"uses_additional_data":false,"paper_date":"2021-05-10","paper":"/paper/you-only-learn-one-representation-unified","paper_url":"https://arxiv.org/abs/2105.04206v1","paper_title":"You Only Learn One Representation: Unified Network for Multiple Tasks","code":"https://github.com/WongKinYiu/yolor","n_code_links":9,"syntology":null},{"rank_in_archive_order":35,"model":"YOLOv9-C","metrics":{"box AP":"53.0"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":36,"model":"PP-YOLOE+_L","metrics":{"FPS (V100, b=1)":"78","box AP":"52.9"},"uses_additional_data":false,"paper_date":"2022-03-30","paper":"/paper/pp-yoloe-an-evolved-version-of-yolo","paper_url":"https://arxiv.org/abs/2203.16250v3","paper_title":"PP-YOLOE: An evolved version of YOLO","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":8,"syntology":{"n_ran":5,"n_unverified":22,"n_samples":27,"n_pointer_only_licence":0}},{"rank_in_archive_order":37,"model":"YOLOv6-L","metrics":{"FPS (V100, b=1)":"98","box AP":"52.8"},"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":38,"model":"RTMDet","metrics":{"box AP":"52.8"},"uses_additional_data":false,"paper_date":"2022-12-14","paper":"/paper/rtmdet-an-empirical-study-of-designing-real","paper_url":"https://arxiv.org/abs/2212.07784v2","paper_title":"RTMDet: An Empirical Study of Designing Real-Time Object Detectors","code":"https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet","n_code_links":14,"syntology":{"n_ran":3,"n_unverified":17,"n_samples":20,"n_pointer_only_licence":0}},{"rank_in_archive_order":39,"model":"DEIM-D-FINE-M","metrics":{"FPS (V100, b=1)":"178 (T4)","box AP":"52.7"},"uses_additional_data":false,"paper_date":"2024-12-05","paper":"/paper/deim-detr-with-improved-matching-for-fast","paper_url":"https://arxiv.org/abs/2412.04234v3","paper_title":"DEIM: DETR with Improved Matching for Fast Convergence","code":"https://github.com/shihuahuang95/deim","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":12,"n_samples":13,"n_pointer_only_licence":13}},{"rank_in_archive_order":40,"model":"YOLOv10-B","metrics":{"box AP":"52.7"},"uses_additional_data":false,"paper_date":"2024-05-23","paper":"/paper/yolov10-real-time-end-to-end-object-detection","paper_url":"https://arxiv.org/abs/2405.14458v2","paper_title":"YOLOv10: Real-Time End-to-End Object Detection","code":"https://github.com/THU-MIG/yolov10","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":41,"model":"YOLOR-P6","metrics":{"FPS (V100, b=1)":"49","box AP":"52.6"},"uses_additional_data":false,"paper_date":"2021-05-10","paper":"/paper/you-only-learn-one-representation-unified","paper_url":"https://arxiv.org/abs/2105.04206v1","paper_title":"You Only Learn One Representation: Unified Network for Multiple Tasks","code":"https://github.com/WongKinYiu/yolor","n_code_links":9,"syntology":null},{"rank_in_archive_order":42,"model":"YOLOv12m","metrics":{"FPS (V100, b=1)":"206 (T4)","box AP":"52.5"},"uses_additional_data":false,"paper_date":"2025-02-20","paper":"/paper/yolov12-a-breakdown-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2502.14740v1","paper_title":"YOLOv12: A Breakdown of the Key Architectural Features","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":43,"model":"PRB-FPN-ELAN","metrics":{"FPS (V100, b=1)":"70","box AP":"52.5"},"uses_additional_data":false,"paper_date":"2020-12-03","paper":"/paper/parallel-residual-bi-fusion-feature-pyramid","paper_url":"https://arxiv.org/abs/2012.01724v5","paper_title":"Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate Single-Shot Object Detection","code":"https://github.com/pingyang1117/PRBNet_PyTorch","n_code_links":1,"syntology":null},{"rank_in_archive_order":44,"model":"GELAN-C","metrics":{"box AP":"52.5"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":45,"model":"D-FINE-M","metrics":{"FPS (V100, b=1)":"178 (T4)","box AP":"52.3"},"uses_additional_data":false,"paper_date":"2024-10-17","paper":"/paper/d-fine-redefine-regression-task-in-detrs-as","paper_url":"https://arxiv.org/abs/2410.13842v1","paper_title":"D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement","code":"https://github.com/huggingface/transformers","n_code_links":5,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":46,"model":"PRB-FPN-CSP","metrics":{"FPS (V100, b=1)":"113","box AP":"51.8"},"uses_additional_data":false,"paper_date":"2020-12-03","paper":"/paper/parallel-residual-bi-fusion-feature-pyramid","paper_url":"https://arxiv.org/abs/2012.01724v5","paper_title":"Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate Single-Shot Object Detection","code":"https://github.com/pingyang1117/PRBNet_PyTorch","n_code_links":1,"syntology":null},{"rank_in_archive_order":47,"model":"YOLOv11m","metrics":{"FPS (V100, b=1)":"212 (T4)","box AP":"51.5"},"uses_additional_data":false,"paper_date":"2024-10-23","paper":"/paper/yolov11-an-overview-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2410.17725v1","paper_title":"YOLOv11: An Overview of the Key Architectural Enhancements","code":"https://github.com/ultralytics/ultralytics","n_code_links":1,"syntology":null},{"rank_in_archive_order":48,"model":"YOLOv9-M","metrics":{"box AP":"51.4"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":49,"model":"YOLOv10-M","metrics":{"box AP":"51.3"},"uses_additional_data":false,"paper_date":"2024-05-23","paper":"/paper/yolov10-real-time-end-to-end-object-detection","paper_url":"https://arxiv.org/abs/2405.14458v2","paper_title":"YOLOv10: Real-Time End-to-End Object Detection","code":"https://github.com/THU-MIG/yolov10","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":50,"model":"MAFYOLOm","metrics":{"box AP":"51.2"},"uses_additional_data":false,"paper_date":"2024-07-05","paper":"/paper/multi-branch-auxiliary-fusion-yolo-with-re","paper_url":"https://arxiv.org/abs/2407.04381v1","paper_title":"Multi-Branch Auxiliary Fusion YOLO with Re-parameterization Heterogeneous Convolutional for accurate object detection","code":"https://github.com/yang-0201/MAF-YOLO","n_code_links":1,"syntology":null},{"rank_in_archive_order":51,"model":"GELAN-M","metrics":{"box AP":"51.1"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":52,"model":"YOLOv3","metrics":{"FPS (V100, b=1)":"123","box AP":"51.0"},"uses_additional_data":false,"paper_date":"2022-03-30","paper":"/paper/pp-yoloe-an-evolved-version-of-yolo","paper_url":"https://arxiv.org/abs/2203.16250v3","paper_title":"PP-YOLOE: An evolved version of YOLO","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":8,"syntology":{"n_ran":5,"n_unverified":22,"n_samples":27,"n_pointer_only_licence":0}},{"rank_in_archive_order":53,"model":"DAMO-YOLO-L","metrics":{"FPS (V100, b=1)":"126","box AP":"50.8"},"uses_additional_data":false,"paper_date":"2022-11-23","paper":"/paper/damo-yolo-a-report-on-real-time-object","paper_url":"https://arxiv.org/abs/2211.15444v4","paper_title":"DAMO-YOLO : A Report on Real-Time Object Detection Design","code":"https://github.com/modelscope/modelscope","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":11,"n_samples":12,"n_pointer_only_licence":12}},{"rank_in_archive_order":54,"model":"D-FINE-S+","metrics":{"FPS (V100, b=1)":"287 (T4)","box AP":"50.7"},"uses_additional_data":true,"paper_date":"2024-10-17","paper":"/paper/d-fine-redefine-regression-task-in-detrs-as","paper_url":"https://arxiv.org/abs/2410.13842v1","paper_title":"D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement","code":"https://github.com/huggingface/transformers","n_code_links":5,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":55,"model":"YOLOv5-X","metrics":{"FPS (V100, b=1)":"62.5","box AP":"50.4"},"uses_additional_data":false,"paper_date":"2021-07-18","paper":"/paper/yolox-exceeding-yolo-series-in-2021","paper_url":"https://arxiv.org/abs/2107.08430v2","paper_title":"YOLOX: Exceeding YOLO Series in 2021","code":"https://github.com/open-mmlab/mmdetection","n_code_links":42,"syntology":{"n_ran":1,"n_unverified":22,"n_samples":23,"n_pointer_only_licence":0}},{"rank_in_archive_order":56,"model":"YOLOv6-S6(1280)","metrics":{"FPS (V100, b=1)":"98","box AP":"50.3"},"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":57,"model":"YOLOv6-M","metrics":{"FPS (V100, b=1)":"175","box AP":"50.0"},"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":58,"model":"PP-YOLOE+_M","metrics":{"box AP":"49.8"},"uses_additional_data":false,"paper_date":"2022-03-30","paper":"/paper/pp-yoloe-an-evolved-version-of-yolo","paper_url":"https://arxiv.org/abs/2203.16250v3","paper_title":"PP-YOLOE: An evolved version of YOLO","code":"https://github.com/PaddlePaddle/PaddleDetection","n_code_links":8,"syntology":{"n_ran":5,"n_unverified":22,"n_samples":27,"n_pointer_only_licence":0}},{"rank_in_archive_order":59,"model":"DAMO-YOLO-M","metrics":{"FPS (V100, b=1)":"233","box AP":"49.2"},"uses_additional_data":false,"paper_date":"2022-11-23","paper":"/paper/damo-yolo-a-report-on-real-time-object","paper_url":"https://arxiv.org/abs/2211.15444v4","paper_title":"DAMO-YOLO : A Report on Real-Time Object Detection Design","code":"https://github.com/modelscope/modelscope","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":11,"n_samples":12,"n_pointer_only_licence":12}},{"rank_in_archive_order":60,"model":"DEIM-D-FINE-S","metrics":{"FPS (V100, b=1)":"287 (T4)","box AP":"49.0"},"uses_additional_data":false,"paper_date":"2024-12-05","paper":"/paper/deim-detr-with-improved-matching-for-fast","paper_url":"https://arxiv.org/abs/2412.04234v3","paper_title":"DEIM: DETR with Improved Matching for Fast Convergence","code":"https://github.com/shihuahuang95/deim","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":12,"n_samples":13,"n_pointer_only_licence":13}},{"rank_in_archive_order":61,"model":"D-FINE-S","metrics":{"FPS (V100, b=1)":"287 (T4)","box AP":"48.5"},"uses_additional_data":false,"paper_date":"2024-10-17","paper":"/paper/d-fine-redefine-regression-task-in-detrs-as","paper_url":"https://arxiv.org/abs/2410.13842v1","paper_title":"D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement","code":"https://github.com/huggingface/transformers","n_code_links":5,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":62,"model":"YOLOv12s","metrics":{"FPS (V100, b=1)":"383 (T4)","box AP":"48.0"},"uses_additional_data":false,"paper_date":"2025-02-20","paper":"/paper/yolov12-a-breakdown-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2502.14740v1","paper_title":"YOLOv12: A Breakdown of the Key Architectural Features","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":63,"model":"MAFYOLOs","metrics":{"box AP":"47.4"},"uses_additional_data":false,"paper_date":"2024-07-05","paper":"/paper/multi-branch-auxiliary-fusion-yolo-with-re","paper_url":"https://arxiv.org/abs/2407.04381v1","paper_title":"Multi-Branch Auxiliary Fusion YOLO with Re-parameterization Heterogeneous Convolutional for accurate object detection","code":"https://github.com/yang-0201/MAF-YOLO","n_code_links":1,"syntology":null},{"rank_in_archive_order":64,"model":"YOLOv11s","metrics":{"FPS (V100, b=1)":"400 (T4)","box AP":"47.0"},"uses_additional_data":false,"paper_date":"2024-10-23","paper":"/paper/yolov11-an-overview-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2410.17725v1","paper_title":"YOLOv11: An Overview of the Key Architectural Enhancements","code":"https://github.com/ultralytics/ultralytics","n_code_links":1,"syntology":null},{"rank_in_archive_order":65,"model":"YOLOv9-S","metrics":{"box AP":"46.8"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":66,"model":"YOLOv10-S","metrics":{"box AP":"46.8"},"uses_additional_data":false,"paper_date":"2024-05-23","paper":"/paper/yolov10-real-time-end-to-end-object-detection","paper_url":"https://arxiv.org/abs/2405.14458v2","paper_title":"YOLOv10: Real-Time End-to-End Object Detection","code":"https://github.com/THU-MIG/yolov10","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":67,"model":"GELAN-S","metrics":{"box AP":"46.7"},"uses_additional_data":false,"paper_date":"2024-02-21","paper":"/paper/yolov9-learning-what-you-want-to-learn-using","paper_url":"https://arxiv.org/abs/2402.13616v2","paper_title":"YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information","code":"https://github.com/ultralytics/ultralytics","n_code_links":5,"syntology":{"n_ran":21,"n_unverified":11,"n_samples":32,"n_pointer_only_licence":11}},{"rank_in_archive_order":68,"model":"DAMO-YOLO-S","metrics":{"FPS (V100, b=1)":"325","box AP":"46"},"uses_additional_data":false,"paper_date":"2022-11-23","paper":"/paper/damo-yolo-a-report-on-real-time-object","paper_url":"https://arxiv.org/abs/2211.15444v4","paper_title":"DAMO-YOLO : A Report on Real-Time Object Detection Design","code":"https://github.com/modelscope/modelscope","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":11,"n_samples":12,"n_pointer_only_licence":12}},{"rank_in_archive_order":69,"model":"YOLOv6-S","metrics":{"FPS (V100, b=1)":"339","box AP":"45.0"},"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":70,"model":"YOLOv4-L","metrics":{"FPS (V100, b=1)":"23","box AP":"43.5"},"uses_additional_data":false,"paper_date":"2020-04-23","paper":"/paper/yolov4-optimal-speed-and-accuracy-of-object","paper_url":"https://arxiv.org/abs/2004.10934v1","paper_title":"YOLOv4: Optimal Speed and Accuracy of Object Detection","code":"https://github.com/tensorflow/models/tree/master/official/vision/beta/projects/yolo","n_code_links":223,"syntology":{"n_ran":24,"n_unverified":160,"n_samples":184,"n_pointer_only_licence":8}},{"rank_in_archive_order":71,"model":"YOLOv4-M","metrics":{"FPS (V100, b=1)":"31","box AP":"43.0"},"uses_additional_data":false,"paper_date":"2020-04-23","paper":"/paper/yolov4-optimal-speed-and-accuracy-of-object","paper_url":"https://arxiv.org/abs/2004.10934v1","paper_title":"YOLOv4: Optimal Speed and Accuracy of Object Detection","code":"https://github.com/tensorflow/models/tree/master/official/vision/beta/projects/yolo","n_code_links":223,"syntology":{"n_ran":24,"n_unverified":160,"n_samples":184,"n_pointer_only_licence":8}},{"rank_in_archive_order":72,"model":"MAFYOLOn","metrics":{"box AP":"42.4"},"uses_additional_data":false,"paper_date":"2024-07-05","paper":"/paper/multi-branch-auxiliary-fusion-yolo-with-re","paper_url":"https://arxiv.org/abs/2407.04381v1","paper_title":"Multi-Branch Auxiliary Fusion YOLO with Re-parameterization Heterogeneous Convolutional for accurate object detection","code":"https://github.com/yang-0201/MAF-YOLO","n_code_links":1,"syntology":null},{"rank_in_archive_order":73,"model":"DAMO-YOLO-T","metrics":{"FPS (V100, b=1)":"397","box AP":"42"},"uses_additional_data":false,"paper_date":"2022-11-23","paper":"/paper/damo-yolo-a-report-on-real-time-object","paper_url":"https://arxiv.org/abs/2211.15444v4","paper_title":"DAMO-YOLO : A Report on Real-Time Object Detection Design","code":"https://github.com/modelscope/modelscope","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":11,"n_samples":12,"n_pointer_only_licence":12}},{"rank_in_archive_order":74,"model":"Faster RCNN-FPN+","metrics":{"FPS (V100, b=1)":"26","box AP":"42"},"uses_additional_data":true,"paper_date":"2020-05-26","paper":"/paper/end-to-end-object-detection-with-transformers","paper_url":"https://arxiv.org/abs/2005.12872v3","paper_title":"End-to-End Object Detection with Transformers","code":"https://github.com/huggingface/transformers","n_code_links":37,"syntology":{"n_ran":59,"n_unverified":33,"n_samples":92,"n_pointer_only_licence":19}},{"rank_in_archive_order":75,"model":"YOLOv4-S","metrics":{"FPS (V100, b=1)":"38","box AP":"41.2"},"uses_additional_data":false,"paper_date":"2020-04-23","paper":"/paper/yolov4-optimal-speed-and-accuracy-of-object","paper_url":"https://arxiv.org/abs/2004.10934v1","paper_title":"YOLOv4: Optimal Speed and Accuracy of Object Detection","code":"https://github.com/tensorflow/models/tree/master/official/vision/beta/projects/yolo","n_code_links":223,"syntology":{"n_ran":24,"n_unverified":160,"n_samples":184,"n_pointer_only_licence":8}},{"rank_in_archive_order":76,"model":"YOLOv12n","metrics":{"FPS (V100, b=1)":"610 (T4)","box AP":"40.6"},"uses_additional_data":false,"paper_date":"2025-02-20","paper":"/paper/yolov12-a-breakdown-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2502.14740v1","paper_title":"YOLOv12: A Breakdown of the Key Architectural Features","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":77,"model":"YOLOv11n","metrics":{"FPS (V100, b=1)":"667 (T4)","box AP":"39.5"},"uses_additional_data":true,"paper_date":"2024-10-23","paper":"/paper/yolov11-an-overview-of-the-key-architectural","paper_url":"https://arxiv.org/abs/2410.17725v1","paper_title":"YOLOv11: An Overview of the Key Architectural Enhancements","code":"https://github.com/ultralytics/ultralytics","n_code_links":1,"syntology":null},{"rank_in_archive_order":78,"model":"YOLOv10-N","metrics":{"box AP":"39.5"},"uses_additional_data":true,"paper_date":"2024-05-23","paper":"/paper/yolov10-real-time-end-to-end-object-detection","paper_url":"https://arxiv.org/abs/2405.14458v2","paper_title":"YOLOv10: Real-Time End-to-End Object Detection","code":"https://github.com/THU-MIG/yolov10","n_code_links":3,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":79,"model":"YOLOv6-N","metrics":{"FPS (V100, b=1)":"779","box AP":"37.5"},"uses_additional_data":true,"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":80,"model":"YOLOv3-L","metrics":{"box AP":"33.0"},"uses_additional_data":true,"paper_date":"2018-04-08","paper":"/paper/yolov3-an-incremental-improvement","paper_url":"http://arxiv.org/abs/1804.02767v1","paper_title":"YOLOv3: An Incremental Improvement","code":"https://github.com/open-mmlab/mmdetection","n_code_links":311,"syntology":{"n_ran":18,"n_unverified":106,"n_samples":124,"n_pointer_only_licence":19}},{"rank_in_archive_order":81,"model":"YOLOv5n","metrics":{"FPS (V100, b=1)":"6.3","box AP":"28.0"},"uses_additional_data":true,"paper_date":"2024-03-22","paper":"/paper/yolov5-6d-advancing-6-dof-instrument-pose","paper_url":"https://ieeexplore.ieee.org/document/10478293","paper_title":"YOLOv5-6D: Advancing 6-DoF Instrument Pose Estimation in Variable X-Ray Imaging Geometries","code":"https://github.com/cviviers/YOLOv5-6D-Pose","n_code_links":1,"syntology":null},{"rank_in_archive_order":82,"model":"DEIM-D-FINE-X+","metrics":{"FPS (V100, b=1)":"78 (T4)"},"uses_additional_data":true,"paper_date":"2024-12-05","paper":"/paper/deim-detr-with-improved-matching-for-fast","paper_url":"https://arxiv.org/abs/2412.04234v3","paper_title":"DEIM: DETR with Improved Matching for Fast Convergence","code":"https://github.com/shihuahuang95/deim","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":12,"n_samples":13,"n_pointer_only_licence":13}}],"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":58,"rows_with_any_sample_ran":52,"distinct_papers_with_graph_line":15,"distinct_papers_with_any_sample_ran":14,"samples_over_distinct_papers":{"n_ran":150,"n_unverified":413,"n_samples":563,"n_pointer_only_licence":86,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":451,"n_unverified":1076,"n_samples":1527,"n_pointer_only_licence":288,"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"}}}