Convolutional neural networks applied to house numbers digit classification

We classify digits of real-world house numbers using convolutional neural networks (ConvNets). Con-vNets are hierarchical feature learning neural networks whose structure is biologically inspired. Unlike many popular vision approaches that are hand-designed, ConvNets can automatically learn a unique set of features optimized for a given task. We augmented the traditional ConvNet architecture by learning multi-stage features and by using Lp pooling and establish a new state-of-the-art of 95.10% accuracy on the SVHN dataset (48% error improvement). Furthermore, we analyze the benefits of different pooling methods and multi-stage features in ConvNets. The source code and a tutorial are available at eblearn.sf.net.

• Publication year

  2012

• Venue

  International Conference on Pattern Recognition

• Publication date

  2012-04-17

• Fields of study

  Computer Science

• Identifiers
  arXiv 1204.3968
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• Source code and a tutorial for the approach are publicly available at eblearn.sf.net.

  Confidence 0.90

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review
• Augmenting the traditional ConvNet architecture with multi-stage features and Lp pooling achieves a new state-of-the-art of 95.10% accuracy on the SVHN dataset, representing a 48% error improvement.

  Confidence 0.95

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review

• convolutional neural networks

  method, architecture

  Hierarchical, biologically inspired feature learning neural networks used here for digit classification.

  Aliases: ConvNets, ConvNet

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review
• eblearn

  software, resource

  An open-source machine learning library providing the source code and tutorial for the described approach.

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review
• house number digit classification

  task

  The task of recognizing and classifying individual digits from images of real-world house numbers.

  Aliases: digit classification

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review
• lp pooling

  method

  A pooling method used to augment the ConvNet architecture for improved feature aggregation.

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review
• multi-stage features

  technique

  Features learned at multiple stages of the network hierarchy rather than a single stage.

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review
• svhn dataset

  dataset, evaluation setting

  A benchmark dataset of real-world house number digit images used for evaluation.

  Aliases: SVHN, Street View House Numbers

  Anonymous (yy3835ars5) extractionAnonymous (12632b8b5f) review

• The German Traffic Sign Recognition Benchmark: A multi-class classification competition

  2011cited by this paper
• A committee of neural networks for traffic sign classification

  2011cited by this paper
• Reading Digits in Natural Images with Unsupervised Feature Learning

  2011influential reference
• Traffic sign recognition with multi-scale Convolutional Networks

  2011influential reference
• Human Tracking Using Convolutional Neural Networks

  2010cited by this paper
• Linear spatial pyramid matching using sparse coding for image classification

  2009cited by this paper
• Character Recognition in Natural Images

  2009cited by this paper
• Learning invariant features through topographic filter maps

  2009cited by this paper
• EBLearn: Open-Source Energy-Based Learning in C++

  2009cited by this paper
• Complex cell pooling and the statistics of natural images

  2007cited by this paper
• The mnist database of handwritten digits

  2005cited by this paper
• Digit classification on signboards for telephone number recognition

  2003cited by this paper
• Gradient-based learning applied to document recognition

  1998cited by this paper
• A model of neuronal responses in visual area MT.

  1998cited by this paper

• Multi-branch local global fusion network for cyclone classification using domain aware pooling

  2026cites this paper
• A systematic review on human action detection and classification architectures using deep learning methodology

  2026cites this paper
• Spiking neural P systems with brain-derived neurotrophic factor

  2026cites this paper
• A comprehensive review of convolutional neural networks: foundations, enhancements and applications

  2026cites this paper
• HR-Mamba: Building Footprint Segmentation with Geometry-Driven Boundary Regularization

  2026cites this paper
• CrowdVLM-R1: Expanding R1 Ability to Vision Language Model for Crowd Counting using Fuzzy Group Relative Policy Reward

  2025cites this paper
• High-Energy Concentration for Federated Learning in Frequency Domain

  2025cites this paper
• LbpRIPooling: A novel rotation-invariant pooling method for CNNs

  2025cites this paper
• Hybrid CNN-BLSTM support vector machine architecture for E2E speech recognition

  2025cites this paper
• ADF-YOLO: lightweight adaptive network for high-speed pharmaceutical blister pack detection

  2025cites this paper
• Binary pooling: A novel approach for local feature extraction in convolutional neural networks

  2025cites this paper
• Effective Memory Management and Sparse Aware Cache for Row-Wise Sparse CNNs

  2025cites this paper
• Distance-informed Neural Processes

  2025cites this paper
• A Parallel Dual-Branch Deep Learning Framework Integrating VMD Feature Engineering and Multiscale Temporal Feature Fusion for Small-Sample Fault Diagnosis

  2025cites this paper
• Energy-Balance-Based Out-of-Distribution Detection of Skin Lesions

  2025cites this paper
• Bridge the Inference Gaps of Neural Processes via Expectation Maximization

  2025influential citation
• A Review of Multi-Source Data Fusion and Analysis Algorithms in Smart City Construction: Facilitating Real Estate Management and Urban Optimization

  2025cites this paper
• Enhancing Neural Network Representations with Prior Knowledge-Based Normalization

  2024cites this paper
• Deep Individual Active Learning: Safeguarding against Out-of-Distribution Challenges in Neural Networks

  2024cites this paper
• Reading and Understanding House Numbers for Delivery Robots Using the “SVHN Dataset”

  2024cites this paper
• UAN: Unsupervised Adaptive Normalization

  2024cites this paper
• Adaptative Context Normalization: A Boost for Deep Learning in Image Processing

  2024cites this paper
• A Deep Learning Model for Predicting User Personality Using Social Media

  2024cites this paper
• Dynamically Choosing the Number of Heads in Multi-Head Attention

  2024cites this paper
• Harnessing Increased Client Participation with Cohort-Parallel Federated Learning

  2024cites this paper
• Automatic fault diagnosis of rolling bearings under multiple working conditions based on unsupervised stack denoising autoencoder

  2024cites this paper
• Deep-Learning and Dynamic Time Warping-Based Approaches for the Diagnosis of Reactor Systems

  2024cites this paper
• SASE: A Searching Architecture for Squeeze and Excitation Operations

  2024cites this paper
• DARE the Extreme: Revisiting Delta-Parameter Pruning For Fine-Tuned Models

  2024cites this paper
• CNN-based noise reduction for multi-channel speech enhancement system with discrete wavelet transform (DWT) preprocessing

  2024cites this paper
• Deep Nets with Subsampling Layers Unwittingly Discard Useful Activations at Test-Time

  2024cites this paper
• Real-time Handwritten Digit Recognition Using CNN on Embedded Systems

  2024influential citation
• UDD: Dataset Distillation via Mining Underutilized Regions

  2024cites this paper
• Vision Transformers for End-to-End Vision-Based Quadrotor Obstacle Avoidance

  2024cites this paper
• Exploiting Inter-sample and Inter-feature Relations in Dataset Distillation

  2024cites this paper
• Enhancing CNN structure and learning through NSGA-II-based multi-objective optimization

  2024cites this paper
• ARMNet: A Network for Image Dimensional Emotion Prediction Based on Affective Region Extraction and Multi-Channel Fusion

  2024cites this paper
• Artificial intelligent systems for vehicle classification: A survey

  2024cites this paper
• A Transfer Learning Based Approach for Human Scar Detection

  2024cites this paper
• Direct Acquisition Optimization for Low-Budget Active Learning

  2024influential citation
• Fault diagnosis method of rolling bearing based on improved deep convolutional neural network

  2024cites this paper
• Understanding ReLU Network Robustness Through Test Set Certification Performance

  2024cites this paper
• Hybrid Quantum–Classical Neural Networks for Efficient MNIST Binary Image Classification

  2024cites this paper
• Context normalization: A new approach for the stability and improvement of neural network performance

  2024cites this paper
• Deep multiscale convolutional feature learning for intracranial hemorrhage classification and weakly supervised localization

  2024cites this paper
• Going Beyond Feature Similarity: Effective Dataset distillation based on Class-aware Conditional Mutual Information

  2024cites this paper
• Continual learning and its industrial applications: A selective review

  2024cites this paper
• Use of Natural Language Processing for the Detection of Hate Speech on Social Media

  2024cites this paper
• Radar Target Recognition and Classification Based on Preprocessing and Improved DenseNet

  2023cites this paper
• RSDiff: remote sensing image generation from text using diffusion model

  2023cites this paper
• MAFNet: A Multiangle Attention Fusion Network for Land Cover Classification

  2023cites this paper
• Self-Attention Parallel Fusion Network for Wind Turbine Gearboxes Fault Diagnosis

  2023cites this paper
• Seismic event classification based on a two-step convolutional neural network

  2023cites this paper
• Scene text understanding: recapitulating the past decade

  2023cites this paper
• A New Global Pooling Method for Deep Neural Networks: Global Average of Top-K Max-Pooling

  2023cites this paper
• CD-CTFM: A Lightweight CNN-Transformer Network for Remote Sensing Cloud Detection Fusing Multiscale Features

  2023cites this paper
• Differentially Private Deep Learning With Dynamic Privacy Budget Allocation and Adaptive Optimization

  2023cites this paper
• Dataset Condensation via Generative Model

  2023cites this paper
• A deep image retrieval network using Max-m-Min pooling and morphological feature generating residual blocks

  2023cites this paper
• Klasifikasi Karakter Tulisan Aksara Jawa Menggunakan Algoritma Convolutional Neural Network

  2023cites this paper
• The SVHN Dataset Is Deceptive for Probabilistic Generative Models Due to a Distribution Mismatch

  2023cites this paper
• A Model Stealing Attack Against Multi-Exit Networks

  2023cites this paper
• Quantification of Uncertainty and Its Applications to Complex Domain for Autonomous Vehicles Perception System

  2023cites this paper
• Bayesian Pseudo-Coresets via Contrastive Divergence

  2023cites this paper
• Context Normalization Layer with Applications

  2023cites this paper
• Hyperspectral Image Classification using Machine Learning Techniques - A Survey

  2023cites this paper
• DiM: Distilling Dataset into Generative Model

  2023cites this paper
• MIM4DD: Mutual Information Maximization for Dataset Distillation

  2023cites this paper
• Deep learning–based long short-term memory recurrent neural networks for monthly rainfall forecasting in Ghana, West Africa

  2023cites this paper
• Trainable Weighted Pooling Method for Text Classification with BERT

  2023cites this paper
• Label Smoothing Loss with Dual-Stream Network Using Separable Convolutional Layers for Retinopathy Grading and Classification

  2023cites this paper
• Diffusion Denoised Smoothing for Certified and Adversarial Robust Out Of Distribution

  2023cites this paper
• SALAD : Source-free Active Label-Agnostic Domain Adaptation for Classification, Segmentation and Detection

  2022cites this paper
• Neighbour feature attention-based pooling

  2022cites this paper
• Improved Alexnet-Based Character Detection Method

  2022influential citation
• HANDWRITTEN MATHEMATICAL EQUATION SOLVER

  2022cites this paper
• On Binding Objects to Symbols Learning Physical Concepts to Understand Real from Fake

  2022cites this paper
• Grey Wolf Optimization Based Hyper-Parameter Optimized Convolution Neural Network for Kidney Image Classification

  2022cites this paper
• CAFE: Learning to Condense Dataset by Aligning Features

  2022cites this paper
• DistillAdapt: Source-Free Active Visual Domain Adaptation

  2022influential citation
• Enabling scalable clinical interpretation of ML-based phenotypes using real world data

  2022cites this paper
• Ghost Expectation Point with Deep Reinforcement Learning in Financial Portfolio Management

  2022influential citation
• CAFE Learning to Condense Dataset by Aligning Features

  2022cites this paper
• An Enhanced Intrusion Detection Model Based on Improved kNN in WSNs

  2022cites this paper
• A Manifold Two-Sample Test Study: Integral Probability Metric with Neural Networks

  2022cites this paper
• An Efficient Heterogeneous SoC with Reconfigurable CNN Accelerator for Edge Processing

  2022cites this paper
• Regressing Image Sub-Population Distributions with Deep Learning

  2022influential citation
• Deep negative correlation classification

  2022cites this paper
• Low-Latency Federated Learning With DNN Partition in Distributed Industrial IoT Networks

  2022cites this paper
• Research on Garbage Image Classification Based on Neural Network

  2022cites this paper
• Optimal Design of Convolutional Neural Network Architectures Using Teaching-Learning-Based Optimization for Image Classification

  2022cites this paper
• Learnable Polyphase Sampling for Shift Invariant and Equivariant Convolutional Networks

  2022influential citation
• Channel attention convolutional recurrent neural network on street view symbol recognition

  2022influential citation
• Convolutional Neural Networks: A Roundup and Benchmark of Their Pooling Layer Variants

  2022cites this paper
• Optimized Hybrid Neural Network for Wind Speed Forecasting

  2022cites this paper
• Protein secondary structure assignment using residual networks

  2022cites this paper
• On the Learnability of Physical Concepts: Can a Neural Network Understand What's Real?

  2022cites this paper
• Saliency Map Based Data Augmentation

  2022cites this paper
• Hyper-flexible Convolutional Neural Networks based on Generalized Lehmer and Power Means

  2022cites this paper
• Building High-throughput Neural Architecture Search Workflows via a Decoupled Fitness Prediction Engine

  2022cites this paper