细粒度图像分类
细粒度图像分类是计算机视觉中的一个任务,旨在将图像划分到更具体的子类别中。该任务要求模型能够识别和区分同一大类别内的细微视觉差异和模式,具有较高的挑战性。其应用价值在于提升图像识别的精度和细致度,适用于生物物种鉴定、商品分类等场景。
Stanford Cars
CUB-200-2011
MetaFormer
(MetaFormer-2,384)
FGVC Aircraft
Inceptionv4
NABirds
HERBS
CUB-200-2011
TBMSL-Net
Oxford 102 Flowers
AutoFormer-S | 384
Stanford Dogs
MP
Oxford-IIIT Pets
µ2Net+ (ViT-L/16)
Caltech-101
Food-101
CAP
Oxford-IIIT Pet Dataset
CompCars
Resnet50 + PMAL
Bird-225
WideResNet-101 (Spinal FC)
Birdsnap
EffNet-L2 (SAM)
SUN397
SEER (RegNet10B - linear eval)
10 Monkey Species
Fruits-360
VGG-19bn
FoodX-251
CSWin-L
Imbalanced CUB-200-2011
PC-Softmax
Kuzushiji-MNIST
BoxCars116K
iNaturalist
TASN
Herbarium 2021 Half–Earth
Herbarium 2022
Conviformer-B
Bottles
CarFlag-1532
CarFlag-563
ResNet101-swp
Con-Text
PHOC descriptor + Fisher Vector Encoding
DIB-10K
MetaFGNet
EMNIST-Digits
VGG-5
EMNIST-Letters
VGG-5
FGVC-Aircraft
EnGraf-Net101 (G=4, H=1)
MNIST
Vanilla FC layer only
QMNIST
VGG-5
SOP
Assemble-ResNet-FGVC-50
STL-10
Pre trained wide-resnet-101