List your company
ResearchFormerly WizWand

Classification on Caltech101 (test)

93.79AccuracyZS_En
55.081265.130675.1885.2294Jun 18, 2014Jan 14, 2016Aug 11, 2017Mar 9, 2019Oct 4, 2020May 2, 2022Nov 28, 2023
Updated 8mo ago

Evaluation results

MethodLinks
ZS_EnEvaluation Protocol=Zero-shot2023.1193.79
SPP-net2014.0693.42
CLIPBackbone=ViT-B/16, Evaluation Protocol=Zero-shot2023.1192.94
MNPsnoise levels=102023.0492.83
MGPnoise levels=102023.0492.68
ReLICv2Backbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=1000 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.0192.2
ReLICv2Backbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=5000 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.0191.4
CLIPBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2023.1191.12
Proposed FrameworkBackbone=ResNet-502018.0591
DnCBackbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=4500 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.0190.2
BYOLBackbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=5000 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.0190.1
CLIPBackbone=RN101, Evaluation Protocol=Zero-shot2023.1189.66
Chatfield et al.2014.0688.54
DeCAF2014.0686.91
Zeiler & Fergus2014.0686.5
Zeiler-Fergus2018.0586.5
CLIPBackbone=RN50, Evaluation Protocol=Zero-shot2023.1185.96
SVMClassifier=SVM, Backbone=seven layers fixed model, Pre-training=ImageNet2018.0585.5
SoftmaxClassifier=Softmax, Backbone=seven layers fixed model, Pre-training=ImageNet2018.0585.4
FL+EN2018.0583.2
DE (LF)noise levels=102023.0481.85
SHDL2018.0581.5
Proposed FrameworkBackbone=VGG-162018.0580.8
FKimplementation=as in Table 6 (a)2014.0677.78
LLCimplementation=as in Table 6 (a)2014.0676.95
Shaban2018.0575.1
VQreported_by=[27]2014.0674.41
DE (EF)noise levels=102023.0473.99
MCDnoise levels=102023.0473.45
TMCnoise levels=102023.0471.77
Image Codes2018.0571.4
ETPnoise levels=102023.0466.62
SNGPnoise levels=102023.0456.57