List your company

Image Classification on Describable Textures Dataset (DTD)

89AccuracyLaCLIP
42.7254.73566.7578.765Jun 17, 2021Oct 13, 2021Feb 9, 2022Jun 8, 2022Oct 5, 2022Feb 1, 2023May 31, 2023

Evaluation results

MethodLinks
LaCLIPArchitecture=ViT-B/16, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0589
CLIPArchitecture=ViT-B/16, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0588.7
LaCLIPArchitecture=ViT-B/32, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0587.5
CLIPArchitecture=ViT-B/32, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0587.4
LaSLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0586.7
SLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0585.8
LaCLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0584.9
CLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0583.3
CLIPArchitecture=ViT-B/16, Pre-training Data=RedCaps, Protocol=5-way 5-shot2023.0582.6
LaCLIPArchitecture=ViT-B/16, Pre-training Data=RedCaps, Protocol=5-way 5-shot2023.0582.4
NLEEPTop-k=k=32022.0778.14
SFDAcomTop-k=k=32022.0778.14
EsViTBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0678.1
LogMETop-k=k=32022.0777.68
SFDATop-k=k=32022.0777.68
NLEEPTop-k=k=22022.0777.25
LogMETop-k=k=22022.0777.14
SFDATop-k=k=22022.0776.95
SFDAcomTop-k=k=22022.0776.95
CLIPBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0676.4
RényiCLEvaluation Protocol=linear evaluation, Multi-crop Augmentation=true2022.0876.3
LaCLIPArchitecture=ViT-B/16, Pre-training Data=CC3M, Protocol=5-way 5-shot2023.0575.8
CLSAEvaluation Protocol=linear evaluation, Multi-crop Augmentation=true2022.0875.5
InfoMinEvaluation Protocol=linear evaluation, Multi-crop Augmentation=false2022.0875
CLSAEvaluation Protocol=linear evaluation, Multi-crop Augmentation=false2022.0875
RényiCLEvaluation Protocol=linear evaluation, Multi-crop Augmentation=false2022.0874.6
CLIPArchitecture=ViT-B/16, Pre-training Data=CC3M, Protocol=5-way 5-shot2023.0574.4
ViT-g/14N-shot=10, Probe=linear regression2023.0574.1
xCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.1074
SupervisedBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0673.1
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe, Reproduced=true2021.0672.6
SoViT-400m/14N-shot=10, Probe=linear regression2023.0572.5
CLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.1072.4
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0672.3
ViT-L/16N-shot=10, Probe=linear regression2023.0572
nCLIPLinear probing=true, Backbone=ViT-B/16, Pre-train=IT35M2022.1070.6
VP (white-box)Access=white-box, Shots=16, Repeated runs=32023.0361.9
BARShots=16, Repeated runs=32023.0347
BlackVIPShots=16, Repeated runs=32023.0345.2
ZSMode=Zero-shot2023.0344.7
VP w/ SPSA-GCShots=16, Repeated runs=3, Optimizer=SPSA-GC2023.0344.5