CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
This addresses the need for more efficient and effective regularization in image classification and localization for computer vision researchers and practitioners, representing a novel method rather than an incremental improvement.
The paper tackles the problem of information loss in regional dropout strategies for training convolutional neural networks by proposing CutMix, a data augmentation method that cuts and pastes patches between images and mixes labels proportionally. The result is consistent state-of-the-art performance on CIFAR and ImageNet classification, with gains in weakly-supervised localization, detection, and captioning tasks.
Regional dropout strategies have been proposed to enhance the performance of convolutional neural network classifiers. They have proved to be effective for guiding the model to attend on less discriminative parts of objects (e.g. leg as opposed to head of a person), thereby letting the network generalize better and have better object localization capabilities. On the other hand, current methods for regional dropout remove informative pixels on training images by overlaying a patch of either black pixels or random noise. Such removal is not desirable because it leads to information loss and inefficiency during training. We therefore propose the CutMix augmentation strategy: patches are cut and pasted among training images where the ground truth labels are also mixed proportionally to the area of the patches. By making efficient use of training pixels and retaining the regularization effect of regional dropout, CutMix consistently outperforms the state-of-the-art augmentation strategies on CIFAR and ImageNet classification tasks, as well as on the ImageNet weakly-supervised localization task. Moreover, unlike previous augmentation methods, our CutMix-trained ImageNet classifier, when used as a pretrained model, results in consistent performance gains in Pascal detection and MS-COCO image captioning benchmarks. We also show that CutMix improves the model robustness against input corruptions and its out-of-distribution detection performances. Source code and pretrained models are available at https://github.com/clovaai/CutMix-PyTorch .