List your company
ResearchFormerly WizWand

Few-shot Image Classification on FC100 (test)

90.57AccuracyMPA
29.542845.386461.2377.0736Jul 21, 2020Jun 24, 2021May 28, 2022May 1, 2023Apr 3, 2024Mar 7, 2025Feb 9, 2026
Updated 8mo ago

Evaluation results

MethodLinks
MPAVenue=None, Pre-trained=true, Way=5-way, Shot=5-shot2026.0290.57——
SPMVenue=AAAI-24, Pre-trained=true, Way=5-way, Shot=5-shot2026.0283.78——
HCTransformersVenue=CVPR-22, Pre-trained=false, Way=5-way, Shot=5-shot2026.0266.42——
EA + Masked PSShots=5-shot, Classifier type=Embedding adaptation (EA), Auxiliary data configuration=Masked embeddings, Pruning level=02020.1265.64——
NCC + Masked PSShots=5-shot, Classifier type=Nearest-centroid classifier (NCC), Auxiliary data configuration=Masked embeddings, Pruning level=02020.1265.52——
CPEAVenue=CVPR-23, Pre-trained=false, Way=5-way, Shot=5-shot2026.0265.02——
SemFew-TransVenue=CVPR-24, Pre-trained=true, Way=5-way, Shot=5-shot2026.0265.02——
R2D2shot=10-shot2021.1061.721277.94
SP-CLIPVenue=CVPR-23, Pre-trained=true, Way=5-way, Shot=5-shot2026.0261.55——
MetaOptNet-SVMshot=10-shot2021.1060.529.5477.03
RFSshots=52021.0860——
MetaOptNet-Ridgeshot=10-shot2021.1059.5211.5675.3
MeLashots=52021.0859.5——
EA + Basic PSShots=5-shot, Classifier type=Embedding adaptation (EA), Auxiliary data configuration=Basic embeddings, Pruning level=02020.1258.13——
MCRNet-SVMbackbone=ResNet-12, shots=5-shot, base-learner=SVM2020.0757.8——
NCC + Basic PSShots=5-shot, Classifier type=Nearest-centroid classifier (NCC), Auxiliary data configuration=Basic embeddings, Pruning level=02020.1257.51——
EA BaselineShots=5-shot, Classifier type=Embedding adaptation (EA), Auxiliary data configuration=Baseline (None), Pruning level=02020.1257.46——
Meta-UAFSVenue=IJCAI-21, Pre-trained=false, Way=5-way, Shot=5-shot2026.0257.43——
MCRNet-RRbackbone=ResNet-12, shots=5-shot, base-learner=Ridge Regression2020.0756.6——
NCC BaselineShots=5-shot, Classifier type=Nearest-centroid classifier (NCC), Auxiliary data configuration=Baseline (None), Pruning level=02020.1256.25——
ProtoNetsshot=10-shot2021.1056.159.9372.94
TADAMbackbone=ResNet-12, shots=5-shot2020.0756.1——
TADAMshots=52021.0856.1——
Baseline-SVMbackbone=ResNet-12, shots=5-shot, base-learner=SVM2020.0755.5——
MetaOptNetshots=52021.0855.5——
MTLbackbone=ResNet-12, shots=5-shot, is_end_to_end=false2020.0755.4——
Fine-tuningbackbone=ResNet-12, shots=5-shot2020.0755.39——
Baseline-RRbackbone=ResNet-12, shots=5-shot, base-learner=Ridge Regression2020.0755.3——
EA + Masked PSShots=1-shot, Classifier type=Embedding adaptation (EA), Auxiliary data configuration=Masked embeddings, Pruning level=02020.1254.59——
MetaOptNet-SVMshot=5-shot2021.1054.455.7574.01
R2D2shot=5-shot2021.1054.356.7274.34
NCC + Masked PSShots=1-shot, Classifier type=Nearest-centroid classifier (NCC), Auxiliary data configuration=Masked embeddings, Pruning level=02020.1253.72——
MetaOptNet-Ridgeshot=5-shot2021.1053.25.9772.65
ProtoNetshots=52021.0852.5——
RM + Masked PSShots=5-shot, Classifier type=Relation module (RM), Auxiliary data configuration=Masked embeddings, Pruning level=02020.1252.02——
EA + Basic PSShots=1-shot, Classifier type=Embedding adaptation (EA), Auxiliary data configuration=Basic embeddings, Pruning level=02020.1251.23——
ProtoNetsshot=5-shot2021.1050.495.5770.31
RM BaselineShots=5-shot, Classifier type=Relation module (RM), Auxiliary data configuration=Baseline (None), Pruning level=02020.1250.28——
NCC + Basic PSShots=1-shot, Classifier type=Nearest-centroid classifier (NCC), Auxiliary data configuration=Basic embeddings, Pruning level=02020.1249.51——
RM + Basic PSShots=5-shot, Classifier type=Relation module (RM), Auxiliary data configuration=Basic embeddings, Pruning level=02020.1249.18——
MCshot=10-shot2021.1049.128.3565.27
Prototypical Networksbackbone=4CONV, shots=5-shot2020.0748.6——
MCshot=5-shot2021.1047.125.1766.33
RFSshots=12021.0844.4——
MAMLshot=10-shot2021.1044.310.5164.98
MeLashots=12021.0844——
MTLbackbone=ResNet-12, shots=1-shot, is_end_to_end=false2020.0743.6——
MAMLshot=5-shot2021.1043.587.5161.64
RM + Basic PSShots=1-shot, Classifier type=Relation module (RM), Auxiliary data configuration=Basic embeddings, Pruning level=02020.1243.22——
RM + Masked PSShots=1-shot, Classifier type=Relation module (RM), Auxiliary data configuration=Masked embeddings, Pruning level=02020.1242.04——
EA BaselineShots=1-shot, Classifier type=Embedding adaptation (EA), Auxiliary data configuration=Baseline (None), Pruning level=02020.1241.52——
Baseline-SVMbackbone=ResNet-12, shots=1-shot, base-learner=SVM2020.0741.1——
MetaOptNetshots=12021.0841.1——
MCRNet-SVMbackbone=ResNet-12, shots=1-shot, base-learner=SVM2020.0741——
MCRNet-RRbackbone=ResNet-12, shots=1-shot, base-learner=Ridge Regression2020.0740.7——
Baseline-RRbackbone=ResNet-12, shots=1-shot, base-learner=Ridge Regression2020.0740.5——
NCC BaselineShots=1-shot, Classifier type=Nearest-centroid classifier (NCC), Auxiliary data configuration=Baseline (None), Pruning level=02020.1240.28——
TADAMbackbone=ResNet-12, shots=1-shot2020.0740.1——
TADAMshots=12021.0840.1——
MetaOptNet-Ridgeshot=1-shot2021.1039.135.4461.28
MetaOptNet-SVMshot=1-shot2021.1038.195.2960.14
R2D2shot=1-shot2021.1037.916.1359.67
Fine-tuningbackbone=ResNet-12, shots=1-shot2020.0737.52——
RM BaselineShots=1-shot, Classifier type=Relation module (RM), Auxiliary data configuration=Baseline (None), Pruning level=02020.1237.51——
ProtoNetshots=12021.0837.5——
ProtoNetsshot=1-shot2021.1037.215.2559.75
MCshot=1-shot2021.1036.566.5356.75
Prototypical Networksbackbone=4CONV, shots=1-shot2020.0735.3——
MAMLshot=1-shot2021.1031.897.3250.13