Few Shot Image Classification
Few-Shot Image Classification 是一种计算机视觉任务，旨在利用少量标注样本（通常少于6个）训练机器学习模型，使其能够对新图像进行分类。该任务的目标是通过最小化监督和数据需求，实现模型对新类别的快速识别与分类，从而在有限的数据条件下提升模型的泛化能力。这种技术在实际应用中具有重要价值，特别是在数据获取困难或成本高昂的场景下。
aPY - 0-Shot
TAFE-Net
AWA - 0-Shot
Synthesised Classifier
AWA1 - 0-Shot
AWA2 - 0-Shot
Bongard-HOI
Human (Amateur)
Caltech-256 5-way (1-shot)
Caltech-256 5-way (5-shot)
MergedNet-Concat
Caltech101
PRE
CIFAR-100
CIFAR-FS - 1-Shot Learning
pseudo-shots
CIFAR-FS - 5-Shot Learning
pseudo-shots
CIFAR-FS 5-way (1-shot)
PT+MAP+SF+SOT (transductive)
CIFAR-FS 5-way (5-shot)
PT+MAP+SF+SOT (transductive)
CIFAR100 5-way (1-shot)
CUB-200 - 0-Shot Learning
TAFE-Net
CUB-200-2011 - 0-Shot
Word CNN-RNN (DS-SJE Embedding)
CUB-200-2011 5-way (1-shot)
MATANet
CUB-200-2011 5-way (5-shot)
MATANet
CUB 200 5-way
EASY 3xResNet12 (transductive)
CUB 200 5-way 1-shot
PT+MAP+SF+BPA (transductive)
CUB 200 5-way 5-shot
PT+MAP+SF+SOT (transductive)
CUB 200 50-way (0-shot)
Prototypical Networks
Dirichlet CUB-200 (5-way, 1-shot)
Dirichlet CUB-200 (5-way, 5-shot)
Dirichlet Mini-Imagenet (5-way, 1-shot)
alpha-TIM
Dirichlet Mini-Imagenet (5-way, 5-shot)
alpha-TIM
Dirichlet Tiered-Imagenet (5-way, 1-shot)
Dirichlet Tiered-Imagenet (5-way, 5-shot)
alpha-TIM
FC100 5-way (1-shot)
R2-D2+Task Aug
FC100 5-way (10-shot)
MTL
FC100 5-way (5-shot)
Fewshot-CIFAR100 - 1-Shot Learning
pseudo-shots
Fewshot-CIFAR100 - 5-Shot Learning
pseudo-shots
Flowers-102 - 0-Shot
Word CNN-RNN (DS-SJE Embedding)
ImageNet - 0-Shot
DebiasPL (ResNet50)
ImageNet (1-shot)
ImageNet - 1-shot
ViT-MoE-15B (Every-2)
ImageNet - 10-shot
MAWS (ViT-6.5B)
ImageNet - 5-shot
ViT-MoE-15B (Every-2)
ImageNet-FS (1-shot, all)
ImageNet-FS (1-shot, novel)
ImageNet-FS (10-shot, all)
KGTN (ResNet-50)
ImageNet-FS (10-shot, novel)
ImageNet-FS (2-shot, all)
ImageNet-FS (2-shot, novel)
ImageNet-FS (5-shot, all)
KGTN-ens (ResNet-50, h+g, max)
ImageNet-FS (5-shot, novel)
iNaturalist 2018 - 1-shot
iNaturalist 2018 - 10-shot
iNaturalist 2018 - 5-shot
iNaturalist (227-way multi-shot)
LaplacianShot
Meta-Dataset
SMAT (DINO-VIT-Base-16-224)
Meta-Dataset Rank
URT
Mini-ImageNet - 1-Shot Learning
PT+MAP
Mini-Imagenet 10-way (1-shot)
Transductive CNAPS + FETI
Mini-Imagenet 10-way (5-shot)
Transductive CNAPS + FETI
mini-ImageNet - 100-Way
GCR
Mini-Imagenet 20-way (1-shot)
TIM-GD
Mini-Imagenet 20-way (5-shot)
TIM-GD
Mini-Imagenet 5-way (1-shot)
PT+MAP+SF+BPA (transductive)
Mini-Imagenet 5-way (10-shot)
PT+MAP
Mini-Imagenet 5-way (5-shot)
CAML [Laion-2b]
Mini-ImageNet-CUB 5-way (1-shot)
PT+MAP
Mini-ImageNet-CUB 5-way (5-shot)
TRIDENT
Mini-ImageNet to CUB - 5 shot learning
TIM-GD
miniImagenet → CUB (5-way 1-shot)
LaplacianShot
miniImagenet → CUB (5-way 5-shot)
LaplacianShot
OMNIGLOT - 1-Shot, 1000 way
APL
OMNIGLOT - 1-Shot, 20-way
GCR
OMNIGLOT - 1-Shot, 423 way
APL
OMNIGLOT - 1-Shot, 5-way
MC2+
OMNIGLOT - 5-Shot, 1000 way
OMNIGLOT - 5-Shot, 20-way
MC2+
OMNIGLOT - 5-Shot, 423 way
APL
OMNIGLOT - 5-Shot, 5-way
DCN6-E
OMNIGLOT-EMNIST 5-way (1-shot)
HyperShot
OMNIGLOT-EMNIST 5-way (5-shot)
ORBIT Clean Video Evaluation
SimpleCNAPs + LITE
ORBIT Clutter Video Evaluation
ProtoNetsVideo
Oxford 102 Flower
RS-FSL
Stanford Cars 5-way (1-shot)
MATANet
Stanford Cars 5-way (5-shot)
MATANet
Stanford Dogs 5-way (1-shot)
MML(KL)
Stanford Dogs 5-way (5-shot)
SUN - 0-Shot
Synthesised Classifier
Tiered ImageNet 10-way (1-shot)
Transductive CNAPS + FETI
Tiered ImageNet 10-way (5-shot)
Transductive CNAPS + FETI
Tiered ImageNet 5-way (1-shot)
PT+MAP
Tiered ImageNet 5-way (5-shot)
CAML [Laion-2b]
UT Zappos50K