Vision transformers have been successfully applied to image recognition tasks due to their ability to capture long-range dependencies within an image. However, there are still gaps in both performance and computational cost between transformers and existing convolutional neural networks (CNNs). In this paper, we aim to address this issue and develop a network that can outperform not only the canonical transformers, but also the high-performance convolutional models. We propose a new transformer based hybrid network by taking advantage of transformers to capture long-range dependencies, and of CNNs to model local features. Furthermore, we scale it to obtain a family of models, called CMTs, obtaining much better accuracy and efficiency than previous convolution and transformer based models. In particular, our CMT-S achieves 83.5% top-1 accuracy on ImageNet, while being 14x and 2x smaller on FLOPs than the existing DeiT and EfficientNet, respectively. The proposed CMT-S also generalizes well on CIFAR10 (99.2%), CIFAR100 (91.7%), Flowers (98.7%), and other challenging vision datasets such as COCO (44.3% mAP), with considerably less computational cost.
翻译:视觉变压器成功地应用到图像识别任务中,因为它们有能力在图像中捕捉长距离依赖性;然而,在变压器和现有变动神经网络(CNNs)之间,在性能和计算成本方面仍然存在差距。在本文件中,我们的目标是解决这一问题,并开发一个不仅能够超越罐状变压器,而且能够超越高性能共变动模型的网络。我们建议利用变压器捕获长距离依赖性,并借助CNNs来模拟本地特征,建立一个基于图像识别任务的新变压器的混合网络。此外,我们扩大变压器的规模,以获得一组模型,称为CMTs,比以前以变动和变压器为基础的模型获得更高的精确度和效率。特别是,我们的CMT-S在图像网络上实现了83.5%的顶级-1精确度,而FLOPs在FLOPs上的精确度比现有的DiT和高效网络分别低14x和2x。拟议的CMT-S还概括了CFAR10(99.2%)、CIFAR100(91.7%)、FLO(98.3%)、Flowers(98.3%)、FLorders(4-malalal)和具有相当大的其他高度数据。