List your company

Image Classification on Caltech101

98.37Base AccuracyHPT
88.521291.078193.63596.1919Nov 19, 2021May 6, 2022Oct 22, 2022Apr 9, 2023Sep 24, 2023Mar 11, 2024Aug 27, 2024

Evaluation results

MethodLinks
HPT2024.0898.37—96.65—94.98
IVLPShots=162024.0498.2893.6595.91——
PromptSRC2024.0898.28—96.39—94.58
CoPrompt2024.0898.27—96.55—94.9
HPT++2024.0898.17—96.96—95.78
MetaPrompt2022.1298.1394.5896.32——
CoOpBackbone=ViT-B/16, Context length (M)=4, Shots=162023.0398.1193.5295.76——
ProMetaRShots=162024.0498.1194.2996.16——
PromptSRCShots=162024.0498.194.0396.02——
MetaPrompt2024.0898.1—96.02—94.03
UNIGRAMShots=162024.0498.0795.1196.57——
ProGradBackbone=ViT-B/16, Context length (M)=4, Shots=162023.0398.0293.8995.91——
ProGrad2024.0898.02—95.91—93.89
CoOp2022.129889.8193.73——
CoOpShots=162024.049889.8193.73——
CoOp2024.0898—93.73—89.81
RPOShots=162024.0497.9794.3796.03——
CoCoOpBackbone=ViT-B/16, Context length (M)=4, Shots=162023.0397.9693.8195.84——
CoCoOp2022.1297.9693.8195.84——
CoCoOpShots=162024.0497.9693.8195.84——
CoCoOp2024.0897.96—95.84—93.81
MaPLe2022.1297.7494.3696.02——
MaPLeShots=162024.0497.7494.3696.02——
MaPLe2024.0897.74—96.02—94.36
KgCoOpBackbone=ViT-B/16, Context length (M)=4, Shots=162023.0397.7294.3996.03——
KgCoOp2024.0897.72—96.03—94.39
X-FM_baseLinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0197.1————
CLIPBackbone=ViT-B/162023.0396.849495.4——
CLIP2022.1296.849495.4——
CLIPShots=Zero-shot2024.0496.849495.4——
CLIPmode=zero-shot2024.0896.84—95.4—94
ST-ProtoPNetBackbone=ResNet34, Explanation=Single-Scale2024.0496.24————
ST-ProtoPNetBackbone=ResNet152, Explanation=Single-Scale2024.0496.09————
Deformable ProtoPNetBackbone=ResNet34, Explanation=Single-Scale2024.0495.95————
BaselineBackbone=ResNet152, Explanation=N/A2024.0495.76————
FLAVALinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0195.7————
BaselineBackbone=ResNet34, Explanation=N/A2024.0494.78————
BASIC-LBackbone=BASIC-L, Evaluation Protocol=Zero-shot2021.1194.7————
CLIPLinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0194.7————
UNIGRAM2024.0494.67————
Any-shift PromptingEvaluation Protocol=Cross-dataset transfer evaluation2024.0294.57————
MCPNetBackbone=ResNet152, Explanation=Multi-Scale2024.0494.54————
CoCoOpPrompts=tp+v, Backbone=ViT-16/B, Source Dataset=ImageNet, Shots=162023.0394.43————
CoCoOpEvaluation Protocol=Cross-dataset transfer evaluation2024.0294.43————
CoCoOp2024.0494.43————
Full dataBackbone=ViT-B/32, Evaluation Protocol=Full Data2023.1194.4————
BASIC-MBackbone=BASIC-M, Evaluation Protocol=Zero-shot2021.1194.2————
VP (white-box)Access=white-box, Shots=16, Repeated runs=32023.0394.2————
DaVinciLinear evaluation=true, Model size=Base, Patch size=16*16, Resolution=224*2242023.0194.1————
KgCoOpPrompts=tp, Backbone=ViT-16/B, Source Dataset=ImageNet, Shots=162023.0393.92————
BADGE + PCB(AS)Backbone=ViT-B/32, Active Learning Strategy=BADGE, Integration Method=PCB(AS), Number of Rounds=82023.1193.85————
BARShots=16, Repeated runs=32023.0393.8————
ProMetaR2024.0493.74————
CoOpPrompts=tp, Backbone=ViT-16/B, Source Dataset=ImageNet, Shots=162023.0393.7————
BlackVIPShots=16, Repeated runs=32023.0393.7————
CoOpEvaluation Protocol=Cross-dataset transfer evaluation2024.0293.7————
CoOp2024.0493.7————
BPLEvaluation Protocol=Cross-dataset transfer evaluation2024.0293.67————
Deformable ProtoPNetBackbone=ResNet152, Explanation=Single-Scale2024.0493.65————
PromptSRC2024.0493.6————
MaPLeEvaluation Protocol=Cross-dataset transfer evaluation2024.0293.53————
MaPLe2024.0493.53————
Entropy + PCB(AS)Backbone=ViT-B/32, Active Learning Strategy=Entropy, Integration Method=PCB(AS), Number of Rounds=82023.1193.48————
GraphAdapterSetting=16-shot2023.0993.33————
MCPNetBackbone=ResNet34, Explanation=Multi-Scale2024.0493.32————
ZSMode=Zero-shot2023.0392.9————
TaskResSetting=16-shot2023.0992.9————
CLIPBackbone=ViT-L/14-336, Evaluation Protocol=Zero-shot2021.1192.8————
Tip-Adapter-FSetting=16-shot2023.0992.63————
CLIP-AdapterSetting=16-shot2023.0992.49————
GraphAdapterSetting=8-shot2023.0992.45————
TaskResSetting=8-shot2023.0992.23————
CoOpSetting=16-shot2023.0991.83————
Tip-Adapter-FSetting=8-shot2023.0991.7————
Coreset + PCB(AS)Backbone=ViT-B/32, Active Learning Strategy=Coreset, Integration Method=PCB(AS), Number of Rounds=82023.1191.7————
BASIC-SBackbone=BASIC-S, Evaluation Protocol=Zero-shot2021.1191.6————
ProGradPrompts=tp, Backbone=ViT-16/B, Source Dataset=ImageNet, Shots=162023.0391.52————
CLIP-AdapterSetting=8-shot2023.0991.4————
GraphAdapterSetting=4-shot2023.0990.97————
Tip-Adapter-FSetting=4-shot2023.0990.87————
TaskResSetting=4-shot2023.0990.63————
EVA-02-CLIP-E/14+Zero-shot=true2024.0290.5————
CoOpSetting=8-shot2023.0990.21————
GraphAdapterSetting=2-shot2023.0990.2————
TaskResSetting=2-shot2023.0990.13————
CLIP-AdapterSetting=4-shot2023.0989.98————
EVA-CLIP-18BZero-shot=true2024.0289.8————
Tip-Adapter-FSetting=2-shot2023.0989.61————
EVA-CLIP-8BZero-shot=true2024.0289.6————
CoOpSetting=4-shot2023.0989.55————
RandomBackbone=ViT-B/32, Active Learning Strategy=Random, Number of Rounds=82023.1189.55————
EVA-01-CLIP-g/14+Zero-shot=true2024.0289.5————
InternVL-CZero-shot=true2024.0289.5————
DFN5B-CLIP-H/14Zero-shot=true2024.0289.5————
VP w/ SPSA-GCShots=16, Repeated runs=3, Optimizer=SPSA-GC2023.0389.4————
OpenCLIP-G/14Zero-shot=true2024.0289.4————
CLIP-AdapterSetting=2-shot2023.0989.37————
CLIPBackbone=ViT-B/16, Evaluation Protocol=Zero-shot2021.1189.3————
GraphAdapterSetting=1-shot2023.0988.9————
DFN5B-CLIP-H/14+Zero-shot=true2024.0288.9————