论文标题
SuperMix：监督混合数据增加
SuperMix: Supervising the Mixing Data Augmentation
论文作者
论文摘要
本文介绍了一种称为SuperMix的有监督的混合增强方法，该方法利用了输入图像中的显着区域来构建混合训练样本。 SuperMix旨在获得富含视觉特征的混合图像，并符合逼真的图像先验。为了提高算法的效率，我们开发了牛顿迭代方法的一种变体，在此问题上，$ 65 \ times $ $ $ $。我们通过对对象分类和知识蒸馏的两项任务进行广泛的评估和消融研究来验证超级混合的有效性。在分类任务上，SuperMix提供了与高级增强方法（例如自动仪和兰加尼）相当的性能。特别是，将SuperMix与Randaugment结合使用ImaNet50的Imagenet上的78.2 \％TOP-1精度。在蒸馏任务上，仅使用教师知识混合的图像分类与最先进的蒸馏方法相当的性能。此外，平均而言，将混合图像纳入蒸馏物镜将CIFAR-100和IMAGENET的性能分别提高了3.4 \％和3.1 \％。 {\ it代码可在https://github.com/alldbi/supermix}上获得。
This paper presents a supervised mixing augmentation method termed SuperMix, which exploits the salient regions within input images to construct mixed training samples. SuperMix is designed to obtain mixed images rich in visual features and complying with realistic image priors. To enhance the efficiency of the algorithm, we develop a variant of the Newton iterative method, $65\times$ faster than gradient descent on this problem. We validate the effectiveness of SuperMix through extensive evaluations and ablation studies on two tasks of object classification and knowledge distillation. On the classification task, SuperMix provides comparable performance to the advanced augmentation methods, such as AutoAugment and RandAugment. In particular, combining SuperMix with RandAugment achieves 78.2\% top-1 accuracy on ImageNet with ResNet50. On the distillation task, solely classifying images mixed using the teacher's knowledge achieves comparable performance to the state-of-the-art distillation methods. Furthermore, on average, incorporating mixed images into the distillation objective improves the performance by 3.4\% and 3.1\% on CIFAR-100 and ImageNet, respectively. {\it The code is available at https://github.com/alldbi/SuperMix}.
