TokenMix：重新思考视觉Transformer中用于数据增强的图像混合方法
TokenMix: Rethinking Image Mixing for Data Augmentation in Vision Transformers
AI总结：
本文提出TokenMix数据增强方法，通过在token级别分区混合图像并利用教师模型激活图生成更准确的学习目标，有效提升视觉Transformer的ImageNet分类精度和抗遮挡鲁棒性。
摘要：
CutMix是一种流行的增强技术，常用于训练现代卷积和Transformer视觉网络。它最初被设计用来鼓励卷积神经网络（CNNs）更多地关注图像的全局上下文而不是局部信息，从而极大地提高了CNNs的性能。然而，我们发现它对于自然具有全局感受野的基于Transformer的架构收益有限。在本文中，我们提出了一种新颖的数据增强技术TokenMix，以提高视觉Transformer的性能。TokenMix通过将混合区域划分为多个分离的部分，在token级别混合两张图像。此外，我们表明CutMix中的混合学习目标，即一对真实标签的线性组合，可能不准确，有时甚至违反直觉。为了获得更合适的目标，我们提出根据来自预训练教师模型的两张图像基于内容的神经激活图来分配目标分数，该教师模型不需要具有高性能。在各种视觉Transformer架构上进行大量实验后，我们表明我们提出的TokenMix帮助视觉Transformer关注前景区域以推断类别，并增强其对遮挡的鲁棒性，带来一致的性能提升。值得注意的是，我们将DeiT-T/S/B的ImageNet top-1准确率提高了+1%。此外，TokenMix受益于更长的训练，使用DeiT-S训练400个epoch在ImageNet上达到81.2%的top-1准确率。代码可在https://github.com/Sense-X/TokenMix获取。
英文摘要：
CutMix is a popular augmentation technique commonly used for training modern convolutional and transformer vision networks. It was originally designed to encourage Convolution Neural Networks (CNNs) to focus more on an image's global context instead of local information, which greatly improves the performance of CNNs. However, we found it to have limited benefits for transformer-based architectures that naturally have a global receptive field. In this paper, we propose a novel data augmentation technique TokenMix to improve the performance of vision transformers. TokenMix mixes two images at token level via partitioning the mixing region into multiple separated parts. Besides, we show that the mixed learning target in CutMix, a linear combination of a pair of the ground truth labels, might be inaccurate and sometimes counter-intuitive. To obtain a more suitable target, we propose to assign the target score according to the content-based neural activation maps of the two images from a pre-trained teacher model, which does not need to have high performance. With plenty of experiments on various vision transformer architectures, we show that our proposed TokenMix helps vision transformers focus on the foreground area to infer the classes and enhances their robustness to occlusion, with consistent performance gains. Notably, we improve DeiT-T/S/B with +1% ImageNet top-1 accuracy. Besides, TokenMix enjoys longer training, which achieves 81.2% top-1 accuracy on ImageNet with DeiT-S trained for 400 epochs. Code is available at https://github.com/Sense-X/TokenMix.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
