paper-with-me

CIFAR-100

홈페이지 · 논문 9,045편

The CIFAR-100 dataset (Canadian Institute for Advanced Research, 100 classes) is a subset of the Tiny Images dataset and consists of 60000 32x32 color images. The 100 classes in the CIFAR-100 are grouped into 20 superclasses. There are 600 images per class. Each image comes with a "fine" label (the class to which it belongs) and a "coarse" label (the superclass to which it belongs). There are 500 training images and 100 testing images per class. The criteria for deciding whether an image belongs to a class were as follows: * The class name should be high on the list of likely answers to the question “What is in this picture?” * The image should be photo-realistic. Labelers were instructed to reject line drawings. * The image should contain only one prominent instance of the object to which the class refers. * The object may be partially occluded or seen from an unusual viewpoint as long as its identity is still clear to the labeler. Source: https://www.cs.toronto.edu/~kriz/cifar.html Image Source: https://www.cs.toronto.edu/~kriz/cifar.html

Images Chinese

벤치마크

Long-tail Learning on CIFAR-100-LT (ρ=100) 결과 330개
Image Classification on CIFAR-100 결과 220개
Long-tail Learning on CIFAR-100-LT (ρ=10) 결과 155개
Semi-Supervised Image Classification on cifar-100, 10000 Labels 결과 58개
Semi-Supervised Image Classification on CIFAR-100, 400 Labels 결과 42개
Semi-Supervised Image Classification on CIFAR-100, 2500 Labels 결과 32개
Image Clustering on CIFAR-100 결과 30개
Knowledge Distillation on CIFAR-100 결과 28개
Neural Architecture Search on CIFAR-100 결과 27개
Few-Shot Class-Incremental Learning on CIFAR-100 결과 22개
Anomaly Detection on One-class CIFAR-100 결과 15개
Class Incremental Learning on cifar100 결과 15개
Incremental Learning on CIFAR-100 - 50 classes + 5 steps of 10 classes 결과 15개
Conditional Image Generation on CIFAR-100 결과 14개
Anomaly Detection on Unlabeled CIFAR-10 vs CIFAR-100 결과 13개
Incremental Learning on CIFAR-100 - 50 classes + 10 steps of 5 classes 결과 13개
Classification on CIFAR-100 결과 12개
Continual Learning on Cifar100 (20 tasks) 결과 9개
Image Generation on CIFAR-100 결과 9개
Personalized Federated Learning on CIFAR-100 결과 7개
Open-World Semi-Supervised Learning on CIFAR-100 결과 6개
Incremental Learning on CIFAR-100 - 50 classes + 25 steps of 2 classes 결과 5개
Network Pruning on CIFAR-100 결과 5개
Out-of-Distribution Detection on CIFAR-100 결과 5개
Class Incremental Learning on CIFAR-100 - 50 classes + 10 steps of 5 classes 결과 4개
Class Incremental Learning on CIFAR-100 - 50 classes + 5 steps of 10 classes 결과 4개
Few-Shot Image Classification on CIFAR100 5-way (1-shot) 결과 4개
Provable Adversarial Defense on CIFAR-100 결과 4개
Semi-Supervised Image Classification on CIFAR-100, 5000Labels 결과 4개
Adversarial Defense on CIFAR-100 결과 3개
Data Free Quantization on CIFAR-100 결과 3개
Small Data Image Classification on CIFAR-100, 1000 Labels 결과 3개
Adversarial Attack on CIFAR-100 결과 2개
Adversarial Robustness on CIFAR-100 결과 2개
Bayesian Inference on cifar100 결과 2개
Image Classification on cifar100 결과 2개
Incremental Learning on CIFAR-100 - 50 classes + 50 steps of 1 class 결과 2개
Learning with coarse labels on cifar100 결과 2개
Learning with noisy labels on CIFAR-100 결과 2개
Non-exemplar-based Class Incremental Learning on cifar100 결과 2개
Novel Class Discovery on cifar100 결과 2개
Self-Supervised Learning on CIFAR-100 결과 2개
Self-Supervised Learning on cifar100 결과 2개
Semi-Supervised Image Classification on CIFAR-100, 1000 Labels 결과 2개
Stochastic Optimization on CIFAR-100 결과 2개
Zero-Shot Learning on CIFAR-100 결과 2개
Classifier calibration on CIFAR-100 결과 1개
Out-of-Distribution Detection on cifar100 결과 1개
Transductive Zero-Shot Classification on CIFAR-100 결과 1개
class-incremental learning on cifar100 결과 1개