Real Time Object Detection On Coco

评估指标

FPS (V100, b=1)

box AP

评测结果

各个模型在此基准测试上的表现结果

			Paper Title	Repository
DEIM-D-FINE-X+	78 (T4)	59.5	DEIM: DETR with Improved Matching for Fast Convergence
D-FINE-X+	78 (T4)	59.3	D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement
YOLOv6-L6(1280)	26	57.2	YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
D-FINE-L+	124 (T4)	57.1	0/1 Deep Neural Networks via Block Coordinate Descent	-
PRB-FPN6-E-ELAN(1280)	31	56.9	Parallel Residual Bi-Fusion Feature Pyramid Network for Accurate Single-Shot Object Detection
YOLOv7-E6E(1280)	36	56.8	YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
YOLOv7-D6(1280)	44	56.6	YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
DEIM-D-FINE-X	78 (T4)	56.5	DEIM: DETR with Improved Matching for Fast Convergence
RT-DETR-H(640)	40 (T4)	56.3	DETRs Beat YOLOs on Real-time Object Detection
YOLOv7-E6(1280)	56	56	YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
D-FINE-X	78 (T4)	55.8	D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement
YOLOv9-E	-	55.6	YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information
YOLOR-D6	30	55.4	You Only Learn One Representation: Unified Network for Multiple Tasks
YOLOv12x	85 (T4)	55.2	YOLOv12: A Breakdown of the Key Architectural Features	-
D-FINE-M+	178 (T4)	55.1	D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement
GELAN-E	-	55.0	YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information
YOLOv7-W6(1280)	84	54.9	YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
YOLOR-E6	37	54.8	You Only Learn One Representation: Unified Network for Multiple Tasks
RT-DETR-X	74 (T4)	54.8	DETRs Beat YOLOs on Real-time Object Detection
PP-YOLOE+_X	45	54.7	PP-YOLOE: An evolved version of YOLO

0 of 82 row(s) selected.