ResizeMix: Mixing Data with Preserved Object Information and True Labels

Abstract

Data augmentation is a powerful technique to increase the diversity of data, which can effectively improve the generalization ability of neural networks in image recognition tasks. Recent data mixing based augmentation strategies have achieved great success. Especially, CutMix uses a simple but effective method to improve the classifiers by randomly cropping a patch from one image and pasting it on another image. To further promote the performance of CutMix, a series of works explore to use the saliency information of the image to guide the mixing. We systematically study the importance of the saliency information for mixing data, and find that the saliency information is not so necessary for promoting the augmentation performance. Furthermore, we find that the cutting based data mixing methods carry two problems of label misallocation and object information missing, which cannot be resolved simultaneously. We propose a more effective but very easily implemented method, namely ResizeMix. We mix the data by directly resizing the source image to a small patch and paste it on another image. The obtained patch preserves more substantial object information compared with conventional cut-based methods. ResizeMix shows evident advantages over CutMix and the saliency-guided methods on both image classification and object detection tasks without additional computation cost, which even outperforms most costly search-based automatic augmentation methods.

References in corpus (17)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Improved Regularization of Convolutional Neural Networks with Cutout
• Wide Residual Networks
• DARTS: Differentiable Architecture Search
• MMDetection: Open MMLab Detection Toolbox and Benchmark
• Data Augmentation Generative Adversarial Networks
• DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
• AutoAugment: Learning Augmentation Policies from Data
• DropBlock: A regularization method for convolutional networks
• Shake-Shake regularization
• Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules
• FMix: Enhancing Mixed Sample Data Augmentation
• SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularization
• Puzzle Mix: Exploiting Saliency and Local Statistics for Optimal Mixup
• Augment your batch: better training with larger batches
• DADA: Differentiable Automatic Data Augmentation
• Fast Neural Network Adaptation via Parameter Remapping and Architecture Search

Cited by in corpus (5)

• High-Performance Large-Scale Image Recognition Without Normalization
• Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup
• Local and Global Context-and-Object-part-Aware Superpixel-based Data Augmentation for Deep Visual Recognition
• It Takes Two to Tango: Mixup for Deep Metric Learning
• Adversarial AutoMixup