List your company

Object Detection on COCO standard (10%)

44.6mAPMixPL
2.89613.72324.5535.377Mar 30, 2022Jul 11, 2022Oct 23, 2022Feb 4, 2023May 18, 2023Aug 30, 2023Dec 12, 2023

Evaluation results

MethodLinks
MixPLDetector=DINO2023.1244.6
Semi-DETRDetector=DINO2023.1243.5
Efficient TeacherDetector type=One-stage anchor-based, Backbone=YOLOv5l, FLOPS=109.59G2023.0237.9
MixPLDetector=FCOS2023.1237.3
Dense TeacherDetector type=One-stage anchor-free, FLOPS=200.59G2023.0237.13
Dense TeacherDetector=FCOS2023.1237.1
DSLDetector type=One-stage anchor-free, FLOPS=200.59G2023.0236.22
DSLDetector=FCOS2023.1236.2
PseCoDetector type=Two-stage anchor-based, FLOPS=202.31G2023.0236.06
PseCoThresholding Strategy=Fully automatic2022.1236.06
LabelMatchThreshold=ACT2022.0635.49
LabelMatchDetector type=Two-stage anchor-based, FLOPS=202.31G2023.0235.49
Unbiased Teacher v2Thresholding Strategy=Fully automatic2022.1235.05
ASTODThresholding Strategy=Fully automatic, Iterations=32022.1234.58
Omni-DETR2022.0334.1
Omni-DETRThresholding Strategy=Manual empirical search2022.1234.1
Efficient TeacherDetector type=One-stage anchor-based, FLOPS=169.61G2023.0234.09
Soft TeacherThreshold=0.92022.0634.04
Soft TeacherDetector type=Two-stage anchor-based, FLOPS=202.31G2023.0234.04
Soft TeacherThresholding Strategy=Manual empirical search2022.1234.04
Unbiased Teacher v2Detector type=One-stage anchor-free, FLOPS=200.59G2023.0232.61
Humber teacherDetector type=Two-stage anchor-based, FLOPS=202.31G2023.0231.61
Humble TeacherThresholding Strategy=Fully automatic2022.1231.61
Humble Teacher2022.0331.6
Unbiased Teacher2022.0331.5
Unbiased TeacherThreshold=0.72022.0631.5
Unbiased TeacherDetector type=Two-stage anchor-based, FLOPS=204.13G2023.0231.5
Unbiased TeacherThresholding Strategy=Manual empirical search2022.1231.5
ISMTThreshold=0.92022.0630.53
ISMTThresholding Strategy=Manual empirical search2022.1230.53
Instant TeachingThreshold=0.92022.0630.4
Instant TeachingDetector type=Two-stage anchor-based, FLOPS=202.31G2023.0230.4
Instant TeachingThresholding Strategy=Manual empirical search2022.1230.4
Unbiased Teacher (re-implemented)Detector type=One-stage anchor-based, FLOPS=169.61G2023.0230.34
Supervised†Thresholding Strategy=Baseline / Teacher2022.1229.34
Deformable DETRImplementation=Re-implementation2022.0329.2
STACThreshold=0.92022.0628.64
STACDetector type=Two-stage anchor-based, FLOPS=202.31G2023.0228.64
STACThresholding Strategy=Manual empirical search2022.1228.64
STAC2022.0328.6
Faster R-CNNImplementation=Re-implementation2022.0325.6
Supervised (One-stage anchor-based)Detector type=One-stage anchor-based, FLOPS=169.61G2023.0224.04
Faster R-CNN2022.0323.9
SupervisedThreshold=N/A2022.0623.86
Supervised (Two-stage)Detector type=Two-stage anchor-based, FLOPS=202.31G2023.0223.86
SupervisedThresholding Strategy=Baseline2022.1223.86
Supervised (One-stage anchor-free)Detector type=One-stage anchor-free, FLOPS=200.59G2023.0223.7
CSDThresholding Strategy=Fully automatic2022.1222.46
Omni-DETRAnnotation Budget Strategy (X%B)=80%B2022.0321.5
Omni-DETRAnnotation Budget Strategy (X%B)=50%B2022.0319.5
UFO2Annotation Budget Strategy (X%B)=80%B2022.0314.1
UFO2Annotation Budget Strategy (X%B)=50%B2022.0311.1
Omni-DETRAnnotation Budget Strategy (X%B)=20%B2022.039.1
UFO2Annotation Budget Strategy (X%B)=20%B2022.034.5