List your company
ResearchFormerly WizWand

Image Classification on CIFAR to STL standard (test)

82.6AccuracyDTA
65.460869.910474.3678.8096Jun 16, 2017Feb 13, 2018Oct 14, 2018Jun 14, 2019Feb 11, 2020Oct 11, 2020Jun 11, 2021
Updated 8mo ago

Evaluation results

MethodLinks
DTA2019.1082.6
Proposed MethodTask=Rotation, Location, Flip2019.0982.1
Proposed MethodTask=Rotation2019.0981.2
Co-DA2019.1081.1
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TFA, Confidence Thresholding=true2017.0680.09
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TF2017.0680.08
VADA2019.0980
VADA2019.1080
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TF, Confidence Thresholding=true2017.0679.73
SE*Protocol=MT+CT+TF2019.1079.7
VADABackbone=Instance Normalization2019.0978.3
Source only (baseline)training=Source domain only2019.1078.2
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=minimal, Confidence Thresholding=true2017.0677.53
Source only (VADA & DIRT-T)mode=source only, Backbone=Instance Normalization2019.0977
Π-Model2019.0976.3
Source only (VADA & DIRT-T)mode=source only2019.0976.3
Source only (Our method)mode=source only2019.0975.6
SupSrcTraining Protocol=Train on Source, Augmentation Type=TF2017.0675.2
SupSrcTraining Protocol=Train on Source, Augmentation Type=TFA2017.0675.18
SupSrcTraining Protocol=Train on Source, Augmentation Type=minimal2017.0672.84
TOHAN2021.0672.8
TOHAN2021.0672.8
S+F2021.0672.1
SHOT2021.0671.9
FT2021.0671.5
T+F2021.0671.3
SupTgtTraining Protocol=Train on Target, Augmentation Type=TF2017.0670.98
dkdHTL2021.0670.8
WA2021.0670.6
SupTgtTraining Protocol=Train on Target, Augmentation Type=TFA2017.0670.03
Target only (baseline)training=Target domain only2019.1070
SupTgtTraining Protocol=Train on Target, Augmentation Type=minimal2017.0667.75
DRCN2019.0966.4
DCRNTraining Protocol=Unsupervised Domain Adaptation2017.0666.37
RevGradTraining Protocol=Unsupervised Domain Adaptation2017.0666.12