Random Erasing Data Augmentation

16 Aug 2017  ·  Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, Yi Yang ·

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.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Robust Object Detection Cityscapes Cutout mPC [AP] 15.7 # 7
Person Re-Identification DukeMTMC-reID TriNet + Random Erasing Rank-1 73.0 # 70
mAP 56.6 # 73
Person Re-Identification DukeMTMC-reID SVDNet + Random Erasing Rank-1 79.3 # 65
mAP 62.4 # 68
Image Classification Fashion-MNIST Random Erasing Percentage error 3.65 # 3
Object Detection PASCAL VOC 2007 I+ORE MAP 76.2% # 18