List your company
ResearchFormerly WizWand

Image Classification on FGVC Aircraft

88.01AccuracySFDA
34.231648.193362.15576.1167Jun 17, 2021Apr 3, 2022Jan 18, 2023Nov 5, 2023Aug 21, 2024Jun 7, 2025Mar 25, 2026
Updated 5mo ago

Evaluation results

MethodLinks
SFDATop-k=k=32022.0788.01——
SFDAcomTop-k=k=32022.0788.01——
SFDATop-k=k=22022.0787.71——
SFDAcomTop-k=k=22022.0787.71——
LogMETop-k=k=22022.0787.23——
LogMETop-k=k=32022.0787.23——
NLEEPTop-k=k=32022.0786.98——
LaCLIPArchitecture=ViT-B/16, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0586.1——
NLEEPTop-k=k=22022.0785.17——
LaCLIPArchitecture=ViT-B/32, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0582.2——
CLIPArchitecture=ViT-B/16, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0580.8——
SVD-ViTComponents=SPC, Layer=112026.0280.68——
SVD-ViTComponents=SPC + ID-RSVD, Layer=112026.0279.93——
SVD-ViTComponents=SPC + SSVA, Layer=102026.0279.84——
SVD-ViTComponents=SPC + SSVA, Layer=112026.0279.63——
MLCDPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0779.6——
SVD-ViTComponents=SPC + SSVA, Layer=92026.0279.45——
SVD-ViTComponents=SPC + ID-RSVD, Layer=92026.0279.45——
SVD-ViTComponents=SPC, Layer=102026.0279.27——
CLIPArchitecture=ViT-B/32, Pre-training Data=LAION-400M, Protocol=5-way 5-shot2023.0578.9——
SVD-ViTComponents=SPC + SSVA + ID-RSVD, Layer=102026.0278.82——
SVD-ViTComponents=SPC, Layer=92026.0278.61——
SVD-ViTComponents=SPC + ID-RSVD, Layer=102026.0278.46——
SVD-ViTComponents=SPC + SSVA + ID-RSVD, Layer=112026.0278.1——
ViTCLS=1, Layer=-2026.0277.86——
SVD-ViTComponents=SPC + SSVA + ID-RSVD, Layer=92026.0277.74——
ViTCLS=8, Layer=-2026.0276.27——
UNICOMPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0774.5——
CLIPPre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0769.4——
LaCLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0568.1——
CLIP+Pre-training Data=WIT-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0768——
OpenCLIPPre-training Data=LAION-400M, Backbone=ViT-L/14, Evaluation Protocol=Linear Probe2024.0764.4——
CLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0562——
EsViTBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0661.1——
RényiCLEvaluation Protocol=linear evaluation, Multi-crop Augmentation=true2022.0861.1——
LaSLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0561——
Supervised(IN1K)Pretrain Dataset=1.2M, Backbone=ResNet-50, Evaluation Protocol=End-to-end fine-tuning2022.0460——
CLIP*Image Encoder=ViT-B/16, Pretrain Dataset Size=400M, Evaluation Protocol=Linear probe2022.0459.5——
RényiCLEvaluation Protocol=linear evaluation, Multi-crop Augmentation=false2022.0858.8——
LaCLIPArchitecture=ViT-B/16, Pre-training Data=RedCaps, Protocol=5-way 5-shot2023.0558.8——
PyramidCLIPPretrain Dataset=143M, Backbone=ResNet-50, Evaluation Protocol=End-to-end fine-tuning2022.0458.4——
GalLoPShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0358.3——
CLIP*Pretrain Dataset=400M, Backbone=ResNet-50, Evaluation Protocol=End-to-end fine-tuning2022.0457.8——
SOT-GLPShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0357.6——
AWTShots=162024.0757.08——
PyramidCLIPImage Encoder=ViT-B/16, Pretrain Dataset Size=143M, Evaluation Protocol=Linear probe2022.0456.5——
SLIPArchitecture=ViT-B/16, Pre-training Data=CC12M, Protocol=5-way 5-shot2023.0556.4——
Skip TuningShot=16, Time (s)=962, Mem. (M)=5282024.1255.57——
Partial-LoRAParameter Count=0.31M2025.1254.58——
CLIPArchitecture=ViT-B/16, Pre-training Data=RedCaps, Protocol=5-way 5-shot2023.0554.5——
LoRAParameter Count=0.67M2025.1254.34——
DoRAParameter Count=0.75M2025.1253.41——
AdaLoRAParameter Count=0.67M2025.1252.74——
VeRAParameter Count=0.10M2025.1252.29——
Partial-AdaLoRAParameter Count=0.30M2025.1252.22——
CLIP*Image Encoder=ViT-B/32, Pretrain Dataset Size=400M, Evaluation Protocol=Linear probe2022.0452——
CLSAEvaluation Protocol=linear evaluation, Multi-crop Augmentation=false2022.0851.6——
SupervisedBackbone=Swin-T, Evaluation Protocol=Linear probe2021.0651.5——
LoRA+Parameter Count=0.67M2025.1251.46——
ProSRCShot=16, Time (s)=14403, Mem. (M)=33732024.1250.83——
PromptSRCShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0350.8——
InfoMinEvaluation Protocol=linear evaluation, Multi-crop Augmentation=false2022.0850.5——
PyramidCLIPImage Encoder=ViT-B/32, Pretrain Dataset Size=143M, Evaluation Protocol=Linear probe2022.0450.2——
CLSAEvaluation Protocol=linear evaluation, Multi-crop Augmentation=true2022.0850——
LaCLIPArchitecture=ViT-B/16, Pre-training Data=CC3M, Protocol=5-way 5-shot2023.0549.2——
CLIPBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0649.1——
SwAVEvaluation Protocol=Linear probe, Backbone=ResNet-50, Pre-training Dataset=ImageNet, Multi-crop Augmentation=2 x 160 + 4 x 962022.0949.1——
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe2021.0648.5——
SupervisedBackbone=ResNet-50, Evaluation Protocol=Linear probe, Reproduced=true2021.0648.4——
MaPLeShot=16, Time (s)=14070, Mem. (M)=30612024.1248.4——
MaPLeShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0348.4——
DISSLEvaluation Protocol=Linear probe, Backbone=ResNet-50, Pre-training Dataset=ImageNet, Multi-crop Augmentation=2 x 160 + 4 x 962022.0948.1——
SCANShot=162025.0547.97——
PLOTShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0346.7——
Skip TuningShot=8, Time (s)=508, Mem. (M)=5282024.1246.5——
CLIPArchitecture=ViT-B/16, Pre-training Data=CC3M, Protocol=5-way 5-shot2023.0546.1——
Linear ProbeShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0345.4——
CoOpShots=162024.0743.4——
CoOpShot=16, Time (s)=3879, Mem. (M)=45512024.1243.4——
CoOpShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0343.4——
ProSRCShot=8, Time (s)=7636, Mem. (M)=33732024.1243.27——
OpenCLIP VIT-H/14zero-shot=true2023.0342.3——
MaPLeShot=8, Time (s)=7060, Mem. (M)=30612024.1242——
SCANShot=82025.0541.4——
StatABackbone=ViT-L/14, Scenario=Separate, Batch Size=1282025.0141.3——
StatABackbone=ViT-L/14, Batch Size=128, Online Test-Time Adaptation=true, Scenario=Separate2025.0141.3——
LoCoOpShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0340.7——
Align+MixShot=16, Backbone=ResNet-502026.0340.5——
GDAShot=16, Backbone=ResNet-502026.0340.3——
SimNLShot=162025.0540.27——
ProDAShots=16, Backbone=ViT-B/16, Pre-training=CLIP2026.0340.2——
Skip TuningShot=4, Time (s)=280, Mem. (M)=5282024.1239.9——
StatABackbone=ViT-L/14, Scenario=High, Batch Size=1282025.0139.6——
StatABackbone=ViT-L/14, Batch Size=128, Online Test-Time Adaptation=true, Scenario=High2025.0139.6——
CoOpShot=8, Time (s)=1988, Mem. (M)=45512024.1239——
StatABackbone=ViT-L/14, Scenario=Medium, Batch Size=1282025.0138.3——
StatABackbone=ViT-L/14, Batch Size=128, Online Test-Time Adaptation=true, Scenario=Medium2025.0138.3——
ProSRCShot=4, Time (s)=3841, Mem. (M)=33732024.1237.47——
CLIPFT + ABackbone=ViT-B/32, Evaluation Protocol=Zero-shot, Text Prompting Strategy=LLM attributes, Training Approach=Fine-tuned2024.0136.41——
TaskResShot=162025.0536.3——