Data to 14 Oct 2026

A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets

arXiv 1707.08819 ↗First posted 27 Jul 2017cs.CV cs.LGcited by 377

Not on the map. Why?

Cited by 34 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The original ImageNet dataset is a popular large-scale benchmark for training Deep Neural Networks. Since the cost of performing experiments (e.g, algorithm design, architecture search, and hyperparameter tuning) on the original dataset might be prohibitive, we propose to consider a downsampled version of ImageNet. In contrast to the CIFAR datasets and earlier downsampled versions of ImageNet, our proposed ImageNet32×32 (and its variants ImageNet64×64 and ImageNet16×16) contains exactly the same number of classes and images as ImageNet, with the only difference that the images are downsampled to 32×32 pixels per image (64×64 and 16×16 pixels for the variants, respectively). Experiments on these downsampled variants are dramatically faster than on the original ImageNet and the characteristics of the downsampled datasets with respect to optimal hyperparameters appear to remain similar. The proposed datasets and scripts to reproduce our results are available at http://image-net.org/download-images and https://github.com/PatrykChrabaszcz/Imagenet32Scripts

Key facts. Cited by 377 papers in the map (data through 14 Oct 2026). First posted on arXiv 27 Jul 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 63; 2022: 54; 2023: 77; 2024: 77; 2025: 66; 2026: 40.

377 papers.

Sort:

By area:

2022Flow Matching for Generative Modelingcited by 5,702
2021LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inferencecited by 475
2023Multisample Flow Matching: Straightening Flows with Minibatch Couplingscited by 278
2022Unlocking High-Accuracy Differentially Private Image Classification through Scalecited by 224
2021Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspectivecited by 169
2024Dynamical Regimes of Diffusion Modelscited by 142
2023Squeeze, Recover and Relabel: Dataset Condensation at ImageNet Scale From A New Perspectivecited by 138
2021Co-Mixup: Saliency Guided Joint Mixup with Supermodular Diversitycited by 128
2022FedALA: Adaptive Local Aggregation for Personalized Federated Learningcited by 122
2022POEM: Out-of-Distribution Detection with Posterior Samplingcited by 108
2022Dataset Condensation with Contrastive Signalscited by 105
2023Elucidating the Exposure Bias in Diffusion Modelscited by 101

38 more

2023Guided Flows for Generative Modeling and Decision Makingcited by 101
2021How Powerful are Performance Predictors in Neural Architecture Search?cited by 90
2023Input Perturbation Reduces Exposure Bias in Diffusion Modelscited by 90
2023Reflected Diffusion Modelscited by 87
2023LLMatic: Neural Architecture Search via Large Language Models and Quality Diversity Optimizationcited by 87
2022Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protectioncited by 86
2023Differentially Private Diffusion Models Generate Useful Synthetic Imagescited by 86
2021HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmarkcited by 84
2022Probing Representation Forgetting in Supervised and Unsupervised Continual Learningcited by 76
2021Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Datacited by 69
2021AutoMix: Unveiling the Power of Mixup for Stronger Classifierscited by 69
2021Eluding Secure Aggregation in Federated Learning via Model Inconsistencycited by 67
2021FlexConv: Continuous Kernel Convolutions with Differentiable Kernel Sizescited by 66
2022Training and Inference on Any-Order Autoregressive Models the Right Waycited by 64
2022Your ViT is Secretly a Hybrid Discriminative-Generative Diffusion Modelcited by 58
2024Flow Matching with General Discrete Paths: A Kinetic-Optimal Perspectivecited by 57
2021Task-Oriented Communication for Multi-Device Cooperative Edge Inferencecited by 52
2021Stable, Fast and Accurate: Kernelized Attention with Relative Positional Encodingcited by 51
2022Rethinking Lipschitz Neural Networks and Certified Robustness: A Boolean Function Perspectivecited by 51
2021Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-Level Backdoor Attackscited by 49
2021Consensus Control for Decentralized Deep Learningcited by 49
2022Boosting Out-of-distribution Detection with Typical Featurescited by 48
2024FedTGP: Trainable Global Prototypes with Adaptive-Margin-Enhanced Contrastive Learning for Data and Model Heterogeneity in Federated Learningcited by 47
2023Maintaining Plasticity in Deep Continual Learningcited by 43
2024How to set AdamW's weight decay as you scale model and dataset sizecited by 43
2022Holomorphic Equilibrium Propagation Computes Exact Gradients Through Finite Size Oscillationscited by 42
2023On Kinetic Optimal Probability Paths for Generative Modelscited by 42
2021CoSSL: Co-Learning of Representation and Classifier for Imbalanced Semi-Supervised Learningcited by 41
2022Test-Time Robust Personalization for Federated Learningcited by 39
2023Eliminating Domain Bias for Federated Learning in Representation Spacecited by 38
2021Adaptive Consistency Regularization for Semi-Supervised Transfer Learningcited by 37
2023Scaling MLPs: A Tale of Inductive Biascited by 37
2023Bespoke Solvers for Generative Flow Modelscited by 37
2021Contrastive Learning Improves Model Robustness Under Label Noisecited by 36
2022FP-Diffusion: Improving Score-based Diffusion Models by Enforcing the Underlying Score Fokker-Planck Equationcited by 36
2023FedCP: Separating Feature Information for Personalized Federated Learning via Conditional Policycited by 36
2023Prototype-Sample Relation Distillation: Towards Replay-Free Continual Learningcited by 35
2024Stochastic Runge-Kutta Methods: Provable Acceleration of Diffusion Modelscited by 35

50 most cited of 377 All 377 (JSON)