Random Erasing Data Augmentation

Abstract

In this paper, we introduce Random Erasing, a new data augmentation method for training the convolutional neural network (CNN). In training, Random Erasing randomly selects a rectangle region in an image and erases its pixels with random values. In this process, training images with various levels of occlusion are generated, which reduces the risk of over-fitting and makes the model robust to occlusion. Random Erasing is parameter learning free, easy to implement, and can be integrated with most of the CNN-based recognition models. Albeit simple, Random Erasing is complementary to commonly used data augmentation techniques such as random cropping and flipping, and yields consistent improvement over strong baselines in image classification, object detection and person re-identification. Code is available at: https://github.com/zhunzhong07/Random-Erasing.

References in corpus (3)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Person Re-identification: Past, Present and Future
• Stochastic Pooling for Regularization of Deep Convolutional Neural Networks

Cited by in corpus (38)

• AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
• Patch-based Output Space Adversarial Learning for Joint Optic Disc and Cup Segmentation
• Bag of Tricks and A Strong Baseline for Deep Person Re-identification
• Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation
• Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges
• Data Augmentation for Object Detection via Progressive and Selective Instance-Switching
• Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-identification
• The Herbarium Challenge 2019 Dataset
• Looking GLAMORous: Vehicle Re-Id in Heterogeneous Cameras Networks with Global and Local Attention
• XNAS: Neural Architecture Search with Expert Advice
• Progressive DARTS: Bridging the Optimization Gap for NAS in the Wild
• Beyond Human Parts: Dual Part-Aligned Representations for Person Re-Identification
• Locality Aware Appearance Metric for Multi-Target Multi-Camera Tracking
• View Confusion Feature Learning for Person Re-identification
• Data Augmentation with Manifold Exploring Geometric Transformations for Increased Performance and Robustness
• Adaptively Connected Neural Networks
• Improving Object Detection with Inverted Attention
• Unity Style Transfer for Person Re-Identification
• Multi-Scale Body-Part Mask Guided Attention for Person Re-identification
• Uncertainty-Aware Multi-Shot Knowledge Distillation for Image-Based Object Re-Identification
• A Preliminary Study on Data Augmentation of Deep Learning for Image Classification
• Backbone Can Not be Trained at Once: Rolling Back to Pre-trained Network for Person Re-Identification
• Style transfer-based image synthesis as an efficient regularization technique in deep learning
• Relation Network for Person Re-identification
• Scalable and Practical Natural Gradient for Large-Scale Deep Learning
• SSAH: Semi-supervised Adversarial Deep Hashing with Self-paced Hard Sample Generation
• Robust, Extensible, and Fast: Teamed Classifiers for Vehicle Tracking and Vehicle Re-ID in Multi-Camera Networks
• Searching for A Robust Neural Architecture in Four GPU Hours
• ROI Regularization for Semi-supervised and Supervised Learning
• Learning Incremental Triplet Margin for Person Re-identification
• Adaptive Deep Metric Embeddings for Person Re-Identification under Occlusions
• VMRFANet:View-Specific Multi-Receptive Field Attention Network for Person Re-identification
• Automated Segmentation of Hip and Thigh Muscles in Metal Artifact-Contaminated CT using Convolutional Neural Network-Enhanced Normalized Metal Artifact Reduction
• Water Supply Prediction Based on Initialized Attention Residual Network
• What Else Can Fool Deep Learning? Addressing Color Constancy Errors on Deep Neural Network Performance
• A Generalization Theory based on Independent and Task-Identically Distributed Assumption
• Unified Multifaceted Feature Learning for Person Re-Identification
• Differentiable Supervector Extraction for Encoding Speaker and Phrase Information in Text Dependent Speaker Verification