SVHN

Street View House Numbers

Dataset Information
Modalities
Images
License
CC
Homepage

Overview

Street View House Numbers (SVHN) is a digit classification benchmark dataset that contains 600,000 32×32 RGB images of printed digits (from 0 to 9) cropped from pictures of house number plates. The cropped images are centered in the digit of interest, but nearby digits and other distractors are kept in the image. SVHN has three sets: training, testing sets and an extra set with 530,000 images that are less difficult and can be used for helping with the training process.

Source: Reading Digits in Natural Images with Unsupervised Feature Learning
Image Source: http://ufldl.stanford.edu/housenumbers/

Variants: svhn,1000, SVHN, 4000 Labels, SVHN, 2000 Labels, SVHN-to-MNIST, SVHN, 500 Labels, SVHN, 40 Labels, SVHN, 250 Labels, SVHN, 1000 labels, SVHN

Associated Benchmarks

This dataset is used in 3 benchmarks:

Recent Benchmark Submissions

Task Model Paper Date
Anomaly Detection RCALAD Spot The Odd One Out: … 2023-04-16
Image Classification Wide-ResNet-28-10 Automatic Data Augmentation via Invariance-Constrained … 2022-09-29
Image Classification ShortNet2-43 Connection Reduction of DenseNet for … 2022-08-02
Image Classification ResNet-18 Benchopt: Reproducible, efficient and collaborative … 2022-06-27
Image Classification WaveMixLite-144/15 WaveMix: A Resource-efficient Neural Network … 2022-05-28
Image Classification EXACT (WRN-16-8) EXACT: How to Train Your … 2022-05-19
Image Classification TripleNet-B Efficient Convolutional Neural Networks on … 2022-04-02
Image Classification SEER (RegNet10B) Vision Models Are More Robust … 2022-02-16
Image Classification Local Mixup LeNet Preventing Manifold Intrusion with Locality: … 2022-01-12
Image Classification ThreshNet79 ThreshNet: An Efficient DenseNet Using … 2022-01-09
Novel Class Discovery AutoNovel AutoNovel: Automatically Discovering and Learning … 2021-06-29
Image Classification PreActResNet18 (AMP) Regularizing Neural Networks via Adversarial … 2020-10-10
Image Classification PyramidNet + AA (AMP) Regularizing Neural Networks via Adversarial … 2020-10-10
Image Classification E2E-M3 Rethinking Recurrent Neural Networks and … 2020-07-30
Image Classification SOPCNN Stochastic Optimization of Plain Convolutional … 2020-01-24
Image Classification EnAET EnAET: A Self-Trained framework for … 2019-11-21
Image Classification PBA [ho2019pba] Population Based Augmentation: Efficient Learning … 2019-05-14
Image Classification MixMatch MixMatch: A Holistic Approach to … 2019-05-06
Image Classification Wide-ResNet-28-10 (Fast AA) Fast AutoAugment 2019-05-01
Image Classification ANODE Augmented Neural ODEs 2019-04-02

Research Papers

Recent papers with results on this dataset: