List your company
ResearchFormerly WizWand

Object Detection on COCO-standard 2% labeled 2017 (train)

28.37mAPListen2Student
12.073216.304120.53524.7659Jun 19, 2022
Updated 8mo ago

Evaluation results

MethodLinks
Listen2StudentBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=32/322022.0628.37
Listen2StudentBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/402022.0628.23
Listen2StudentBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/82022.0626.14
Unbiased TeacherBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=32/322022.0624.3
Instant TeachingBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/82022.0622.45
ISMTBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/82022.0622.43
Unbiased TeacherBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/82022.0622.23
Humble TeacherBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/82022.0621.72
STACBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/82022.0618.25
CSDBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based, Batch Size (labeled/unlabeled)=8/82022.0613.93
SupervisedBackbone=FasterRCNN-ResNet50, Detector Type=Anchor-based2022.0612.7