paper-with-me

Object Detection 벤치마크

Object Detection on COCO 2017 val

165개 결과 · ⬇ CSV · JSON

AP

39.7 45.23 50.75 56.27 61.8 2021-10 2026-09 ViDT Swin-base — 49.2 (2021-10-08) ViDT Swin-small — 47.5 (2021-10-08) ViDT Swin-tiny — 44.8 (2021-10-08) ViDT Swin-nano — 40.4 (2021-10-08) ViDT Swin-base — 49.2 (2021-10-08) ViDT Swin-small — 47.5 (2021-10-08) ViDT Swin-tiny — 44.8 (2021-10-08) ViDT Swin-nano — 40.4 (2021-10-08) ViDT Swin-base — 49.2 (2021-10-08) ViDT Swin-small — 47.5 (2021-10-08) ViDT Swin-tiny — 44.8 (2021-10-08) ViDT Swin-nano — 40.4 (2021-10-08) ViDT Swin-base — 49.2 (2021-10-08) ViDT Swin-small — 47.5 (2021-10-08) ViDT Swin-tiny — 44.8 (2021-10-08) ViDT Swin-nano — 40.4 (2021-10-08) ViDT Swin-base — 49.2 (2021-10-08) ViDT Swin-small — 47.5 (2021-10-08) ViDT Swin-tiny — 44.8 (2021-10-08) ViDT Swin-nano — 40.4 (2021-10-08) RF-ConvNeXt-T Cascade R-CNN — 50.9 (2022-06-14) RF-ConvNeXt-T Cascade R-CNN — 50.9 (2022-06-14) RF-ConvNeXt-T Cascade R-CNN — 50.9 (2022-06-14) RF-ConvNeXt-T Cascade R-CNN — 50.9 (2022-06-14) RF-ConvNeXt-T Cascade R-CNN — 50.9 (2022-06-14) MogaNet-XL (Cascade Mask R-CNN) — 56.2 (2022-11-07) MogaNet-L (Cascade Mask R-CNN) — 53.3 (2022-11-07) MogaNet-B (Cascade Mask R-CNN) — 52.6 (2022-11-07) MogaNet-S (Cascade Mask R-CNN) — 51.6 (2022-11-07) MogaNet-L (Mask R-CNN 1x) — 49.4 (2022-11-07) MogaNet-L (RetinaNet 1x) — 48.7 (2022-11-07) MogaNet-B (Mask R-CNN 1x) — 47.9 (2022-11-07) MogaNet-B (RetinaNet 1x) — 47.7 (2022-11-07) MogaNet-S (Mask R-CNN 1x) — 46.7 (2022-11-07) MogaNet-S (RetinaNet 1x) — 45.8 (2022-11-07) MogaNet-T (Mask R-CNN 1x) — 42.6 (2022-11-07) MogaNet-T (RetinaNet 1x) — 41.4 (2022-11-07) MogaNet-XT (Mask R-CNN 1x) — 40.7 (2022-11-07) MogaNet-XT (RetinaNet 1x) — 39.7 (2022-11-07) MogaNet-XL (Cascade Mask R-CNN) — 56.2 (2022-11-07) MogaNet-L (Cascade Mask R-CNN) — 53.3 (2022-11-07) MogaNet-B (Cascade Mask R-CNN) — 52.6 (2022-11-07) MogaNet-S (Cascade Mask R-CNN) — 51.6 (2022-11-07) MogaNet-L (Mask R-CNN 1x) — 49.4 (2022-11-07) MogaNet-L (RetinaNet 1x) — 48.7 (2022-11-07) MogaNet-B (Mask R-CNN 1x) — 47.9 (2022-11-07) MogaNet-B (RetinaNet 1x) — 47.7 (2022-11-07) MogaNet-S (Mask R-CNN 1x) — 46.7 (2022-11-07) MogaNet-S (RetinaNet 1x) — 45.8 (2022-11-07) MogaNet-T (Mask R-CNN 1x) — 42.6 (2022-11-07) MogaNet-T (RetinaNet 1x) — 41.4 (2022-11-07) MogaNet-XT (Mask R-CNN 1x) — 40.7 (2022-11-07) MogaNet-XT (RetinaNet 1x) — 39.7 (2022-11-07) MogaNet-XL (Cascade Mask R-CNN) — 56.2 (2022-11-07) MogaNet-L (Cascade Mask R-CNN) — 53.3 (2022-11-07) MogaNet-B (Cascade Mask R-CNN) — 52.6 (2022-11-07) MogaNet-S (Cascade Mask R-CNN) — 51.6 (2022-11-07) MogaNet-L (Mask R-CNN 1x) — 49.4 (2022-11-07) MogaNet-L (RetinaNet 1x) — 48.7 (2022-11-07) MogaNet-B (Mask R-CNN 1x) — 47.9 (2022-11-07) MogaNet-B (RetinaNet 1x) — 47.7 (2022-11-07) MogaNet-S (Mask R-CNN 1x) — 46.7 (2022-11-07) MogaNet-S (RetinaNet 1x) — 45.8 (2022-11-07) MogaNet-T (Mask R-CNN 1x) — 42.6 (2022-11-07) MogaNet-T (RetinaNet 1x) — 41.4 (2022-11-07) MogaNet-XT (Mask R-CNN 1x) — 40.7 (2022-11-07) MogaNet-XT (RetinaNet 1x) — 39.7 (2022-11-07) MogaNet-XL (Cascade Mask R-CNN) — 56.2 (2022-11-07) MogaNet-L (Cascade Mask R-CNN) — 53.3 (2022-11-07) MogaNet-B (Cascade Mask R-CNN) — 52.6 (2022-11-07) MogaNet-S (Cascade Mask R-CNN) — 51.6 (2022-11-07) MogaNet-L (Mask R-CNN 1x) — 49.4 (2022-11-07) MogaNet-L (RetinaNet 1x) — 48.7 (2022-11-07) MogaNet-B (Mask R-CNN 1x) — 47.9 (2022-11-07) MogaNet-B (RetinaNet 1x) — 47.7 (2022-11-07) MogaNet-S (Mask R-CNN 1x) — 46.7 (2022-11-07) MogaNet-S (RetinaNet 1x) — 45.8 (2022-11-07) MogaNet-T (Mask R-CNN 1x) — 42.6 (2022-11-07) MogaNet-T (RetinaNet 1x) — 41.4 (2022-11-07) MogaNet-XT (Mask R-CNN 1x) — 40.7 (2022-11-07) MogaNet-XT (RetinaNet 1x) — 39.7 (2022-11-07) MogaNet-XL (Cascade Mask R-CNN) — 56.2 (2022-11-07) MogaNet-L (Cascade Mask R-CNN) — 53.3 (2022-11-07) MogaNet-B (Cascade Mask R-CNN) — 52.6 (2022-11-07) MogaNet-S (Cascade Mask R-CNN) — 51.6 (2022-11-07) MogaNet-L (Mask R-CNN 1x) — 49.4 (2022-11-07) MogaNet-L (RetinaNet 1x) — 48.7 (2022-11-07) MogaNet-B (Mask R-CNN 1x) — 47.9 (2022-11-07) MogaNet-B (RetinaNet 1x) — 47.7 (2022-11-07) MogaNet-S (Mask R-CNN 1x) — 46.7 (2022-11-07) MogaNet-S (RetinaNet 1x) — 45.8 (2022-11-07) MogaNet-T (Mask R-CNN 1x) — 42.6 (2022-11-07) MogaNet-T (RetinaNet 1x) — 41.4 (2022-11-07) MogaNet-XT (Mask R-CNN 1x) — 40.7 (2022-11-07) MogaNet-XT (RetinaNet 1x) — 39.7 (2022-11-07) SQR-Adamixer-R101 — 49.8 (2022-12-15) SQR-Adamixer-R50 — 48.9 (2022-12-15) SQR-Adamixer-R101 — 49.8 (2022-12-15) SQR-Adamixer-R50 — 48.9 (2022-12-15) SQR-Adamixer-R101 — 49.8 (2022-12-15) SQR-Adamixer-R50 — 48.9 (2022-12-15) SQR-Adamixer-R101 — 49.8 (2022-12-15) SQR-Adamixer-R50 — 48.9 (2022-12-15) SQR-Adamixer-R101 — 49.8 (2022-12-15) SQR-Adamixer-R50 — 48.9 (2022-12-15) YOLOv6-L6(46 fps, V100, bs1) — 57.2 (2023-01-13) YOLOv6-L6(46 fps, V100, bs1) — 57.2 (2023-01-13) YOLOv6-L6(46 fps, V100, bs1) — 57.2 (2023-01-13) YOLOv6-L6(46 fps, V100, bs1) — 57.2 (2023-01-13) YOLOv6-L6(46 fps, V100, bs1) — 57.2 (2023-01-13) Salience-DETR (Focal-L 1x) — 57.3 (2024-03-24) Salience-DETR (Swin-L 1x) — 56.5 (2024-03-24) Salience-DETR (ResNet50 1x) — 50.0 (2024-03-24) Salience-DETR (Focal-L 1x) — 57.3 (2024-03-24) Salience-DETR (Swin-L 1x) — 56.5 (2024-03-24) Salience-DETR (ResNet50 1x) — 50.0 (2024-03-24) Salience-DETR (Focal-L 1x) — 57.3 (2024-03-24) Salience-DETR (Swin-L 1x) — 56.5 (2024-03-24) Salience-DETR (ResNet50 1x) — 50.0 (2024-03-24) Salience-DETR (Focal-L 1x) — 57.3 (2024-03-24) Salience-DETR (Swin-L 1x) — 56.5 (2024-03-24) Salience-DETR (ResNet50 1x) — 50.0 (2024-03-24) Salience-DETR (Focal-L 1x) — 57.3 (2024-03-24) Salience-DETR (Swin-L 1x) — 56.5 (2024-03-24) Salience-DETR (ResNet50 1x) — 50.0 (2024-03-24) Relation-DETR (Swin-L 2x) — 58.1 (2024-07-16) Relation-DETR (Swin-L 1x) — 57.8 (2024-07-16) Relation-DETR (ResNet50 2x) — 52.1 (2024-07-16) Relation-DETR (ResNet50 1x) — 51.7 (2024-07-16) Relation-DETR (Swin-L 2x) — 58.1 (2024-07-16) Relation-DETR (Swin-L 1x) — 57.8 (2024-07-16) Relation-DETR (ResNet50 2x) — 52.1 (2024-07-16) Relation-DETR (ResNet50 1x) — 51.7 (2024-07-16) Relation-DETR (Swin-L 2x) — 58.1 (2024-07-16) Relation-DETR (Swin-L 1x) — 57.8 (2024-07-16) Relation-DETR (ResNet50 2x) — 52.1 (2024-07-16) Relation-DETR (ResNet50 1x) — 51.7 (2024-07-16) Relation-DETR (Swin-L 2x) — 58.1 (2024-07-16) Relation-DETR (Swin-L 1x) — 57.8 (2024-07-16) Relation-DETR (ResNet50 2x) — 52.1 (2024-07-16) Relation-DETR (ResNet50 1x) — 51.7 (2024-07-16) Relation-DETR (Swin-L 2x) — 58.1 (2024-07-16) Relation-DETR (Swin-L 1x) — 57.8 (2024-07-16) Relation-DETR (ResNet50 2x) — 52.1 (2024-07-16) Relation-DETR (ResNet50 1x) — 51.7 (2024-07-16) Mr. DETR (Swin-L, 1x, 5cale) — 61.8 (2024-12-13) Mr. DETR (Swin-L, 1x, 4scale) — 58.4 (2024-12-13) Mr. DETR (Swin-L, 1x, 5cale) — 61.8 (2024-12-13) Mr. DETR (Swin-L, 1x, 4scale) — 58.4 (2024-12-13) Mr. DETR (Swin-L, 1x, 5cale) — 61.8 (2024-12-13) Mr. DETR (Swin-L, 1x, 4scale) — 58.4 (2024-12-13) Mr. DETR (Swin-L, 1x, 5cale) — 61.8 (2024-12-13) Mr. DETR (Swin-L, 1x, 4scale) — 58.4 (2024-12-13) Mr. DETR (Swin-L, 1x, 5cale) — 61.8 (2024-12-13) Mr. DETR (Swin-L, 1x, 4scale) — 58.4 (2024-12-13) MI-DETR (Swin-L 1x) — 58.2 (2025-03-03) MI-DETR (Swin-L 1x) — 58.2 (2025-03-03) MI-DETR (Swin-L 1x) — 58.2 (2025-03-03) MI-DETR (Swin-L 1x) — 58.2 (2025-03-03) MI-DETR (Swin-L 1x) — 58.2 (2025-03-03) ViDT Swin-base — 49.2 (2021-10-08) RF-ConvNeXt-T Cascade R-CNN — 50.9 (2022-06-14) MogaNet-XL (Cascade Mask R-CNN) — 56.2 (2022-11-07) YOLOv6-L6(46 fps, V100, bs1) — 57.2 (2023-01-13) Salience-DETR (Focal-L 1x) — 57.3 (2024-03-24) Relation-DETR (Swin-L 2x) — 58.1 (2024-07-16) Mr. DETR (Swin-L, 1x, 5cale) — 61.8 (2024-12-13)
RankModel APAP50AP75APLAPMAPSParam. Extra Training Data PaperCodeYear
1 Mr. DETR (Swin-L, 1x, 5cale) 61.879.067.675.765.647.7– ✓ Mr. DETR: Instructive Multi-Route Training for Detection Transformers Visual-AI/Mr.DETR 2024
2 Mr. DETR (Swin-L, 1x, 4scale) 58.476.363.975.362.840.8– Mr. DETR: Instructive Multi-Route Training for Detection Transformers Visual-AI/Mr.DETR 2024
3 MI-DETR (Swin-L 1x) 58.276.563.474.662.842.5– MI-DETR: An Object Detection Model with Multi-time Inquiries Mechanism CQU-ADHRI-Lab/MI-DETR 2025
4 Relation-DETR (Swin-L 2x) 58.176.463.573.563.041.8– Relation DETR: Exploring Explicit Position Relation Prior for Object Detection xiuqhou/relation-detr · xiuqhou/Salience-DETR 2024
5 Relation-DETR (Swin-L 1x) 57.876.162.974.462.141.2– Relation DETR: Exploring Explicit Position Relation Prior for Object Detection xiuqhou/relation-detr · xiuqhou/Salience-DETR 2024
6 Salience-DETR (Focal-L 1x) 57.375.562.374.561.840.9220M Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement xiuqhou/relation-detr · xiuqhou/Salience-DETR · xunull/read-Salience-DETR 2024
7 YOLOv6-L6(46 fps, V100, bs1) 57.274.5––––– YOLOv6 v3.0: A Full-Scale Reloading PaddlePaddle/PaddleDetection · meituan/yolov6 · PaddlePaddle/PaddleYOLO · +2 2023
8 Salience-DETR (Swin-L 1x) 56.575.061.572.861.240.2210M Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement xiuqhou/relation-detr · xiuqhou/Salience-DETR · xunull/read-Salience-DETR 2024
9 MogaNet-XL (Cascade Mask R-CNN) 56.2–––––– MogaNet: Multi-order Gated Aggregation Network chengtan9907/OpenSTL · chengtan9907/simvpv2 · Westlake-AI/openmixup · +4 2022
10 MogaNet-L (Cascade Mask R-CNN) 53.3–––––– MogaNet: Multi-order Gated Aggregation Network chengtan9907/OpenSTL · chengtan9907/simvpv2 · Westlake-AI/openmixup · +4 2022
1–10 / 165 다음 → 페이지당 10 20 50 100