List your company
ResearchFormerly WizWand

Object Detection on MS-COCO 1% labeled protocol

23.86mAPVirtual Category Learning (with tricks)
8.457612.456316.45520.4537May 10, 2020Sep 20, 2020Jan 31, 2021Jun 13, 2021Oct 24, 2021Mar 6, 2022Jul 17, 2022
Updated 8mo ago

Evaluation results

MethodLinks
Virtual Category Learning (with tricks)Configuration=Training tricks in Soft Teacher2022.0723.86
Dual Pseudo-label Polishing Framework2022.0723.55
Virtual Category LearningConfiguration=Same settings as Unbiased Teacher2022.0721.61
Unbiased Teacher2022.0720.75
Unbiased TeacherRemark=ICLR 20212022.0720.75
Soft Teacher2022.0720.46
Soft TeacherRemark=ICCV 20212022.0720.46
Interactive2022.0718.88
Region Uncertainty QuantificationStrong data augmentation=true2021.1118.41
Combating Noise2022.0718.41
Instant-Teaching*Backbone=R50-FPN, co-rectify scheme=true2021.0318.05
Instant-TeachingStrong data augmentation=true2021.1118.05
Instant Teaching2022.0718.05
Instant-Teaching*Remark=CVPR 2021, Note=Asterisk denotes variation mentioned in context2022.0718.05
Unbiased TeacherStrong data augmentation=true2021.1117.84
Humble TeacherStrong data augmentation=true2021.1116.96
Humble Teacher2022.0716.96
Humble TeacherRemark=CVPR 20212022.0716.96
Instant-TeachingBackbone=R50-FPN, co-rectify scheme=false2021.0316
Instant-TeachingRemark=CVPR 20212022.0716
STACBackbone=ResNet-50 with FPN, Augmentation=C+{B,G}+Cutout, Training Data=Semi-supervised2020.0513.97
STACBackbone=R50-FPN2021.0313.97
STACStrong data augmentation=true2021.1113.97
STAC2022.0713.97
STACRemark=arxiv 20202022.0713.97
CSDStrong data augmentation=false2021.1110.51
CSD2022.0710.51
CSD†Backbone=R50-FPN2021.0310.2
Supervised2022.0710
Supervised†Backbone=ResNet-50 with FPN, Augmentation=C+{B,G}+Cutout, Training Data=Labeled data only2020.059.83
SupervisedBackbone=ResNet-50 with FPN, Augmentation=Standard, Training Data=Labeled data only2020.059.05
SupervisedBackbone=R50-FPN2021.039.05
Supervised2022.079.05