Papers with Code · CC BY-SA 4.0
Image Classification
Image Classification is a fundamental task in vision recognition that aims to understand and categorize an image as a whole under a specific label. Unlike object detection, which involves classification and location of multiple objects within an image, image classification typically pertains to single-object images. When the classification becomes highly detailed or reaches instance-level, it is often referred to as image retrieval, which also involves finding similar images in a large database.
<span class="description-source">Source: Metamorphic Testing for Object Detection Systems </span>
데이터셋
505개Bongard-HOICIFAR-10CIFAR-10, 1000 LabelsCINIC-10Clothing1M (using clean data)Dirichlet Tiered-Imagenet (5-way, 1-shot)Dirichlet Tiered-Imagenet (5-way, 5-shot)ImageNetSTL-10CIFAR-10-LT (ρ=50)Dirichlet CUB-200 (5-way, 1-shot)Dirichlet CUB-200 (5-way, 5-shot)GasHisSDBImageNet - 1-shotImageNet - 5-shotImageNet-FS (2-shot, novel)ImageNet-FS (5-shot, all)Mini-ImageNet-CUB 5-way (5-shot)CompCarsEMNIST-DigitsImageNet - 10-shotImageNet-FS (1-shot, novel)Places365Tiered ImageNet 5-way (5-shot)noise padded CIFAR-10smallNORB75 Superpixel MNISTBigEarthNet (official test set)CIFAR-10, 500 LabelsColored-MNIST(with spurious correlation)