120 datasets found

The AR face database contains 100 classes of faces with 26 face images per class with various natural variation and occlusions.

The dataset used in the paper is not explicitly mentioned, but it is implied to be ImageNet-R/A/Sk for ImageNet-R/A/Sk classification.

The dataset used in the paper is not explicitly mentioned, but it is implied to be ImageNet-C/C for ImageNet-C/C classification.

The dataset used in the paper is not explicitly mentioned, but it is implied to be ImageNet-50/100/200 for ImageNet-50/100/200 classification.

The ImageNet-32 dataset is a subset of the ImageNet dataset, containing 1,281,167 training samples and 50,000 test samples, distributed across 1,000 labels.

The dataset used in the paper is OxfordPets, a dataset of 33 animal categories.

The dataset used in the paper is the Cars Overhead With Context (COWC) dataset, which contains images of cars in overhead imagery.

MobileNets, MnasNets, EfficientNets, and ResNets

Long-tailed classification has been extensively studied in recent years due to its importance in real-world applications with heavily imbalanced data distribution.

Four remote sensing datasets were used for classification with both small and large training samples.

Seasonal contrast: Unsupervised pre-training from uncurated remote sensing data.

Remote sensing image scene classification with self-supervised paradigm under limited labeled samples.

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used the ImageNet dataset and a subset of the CC12M dataset for training.

The miniImagenet and tieredImageNet datasets are used for few-shot classification experiments.

Image classification results on CIFAR-10 and CIFAR-100 benchmarks suggested that our new network architecture performs better than ResNet.

The dataset used in the paper is not explicitly described, but it is mentioned that the authors pre-trained GANs on four datasets: MNIST, Fashion-MNIST, CIFAR-10, and CelebA.

Supervised learning of deep neural networks heavily relies on large-scale datasets annotated by high-quality labels. In contrast, mislabeled samples can significantly degrade...

The dataset used in the paper is not explicitly described. However, it is mentioned that the authors used various benchmarks such as ImageNet, ImageNet ReaL, ImageNet V2, etc.

The dataset used for evaluation of the proposed method. It contains images of humans and faces, with pre-annotated bounding boxes for persons and faces.

The dataset used in the paper is the ImageNet and MS COCO benchmarks.