Image and Model Transformation with Secret Key for Vision Transformer
Image and Model Transformation with Secret Key for Vision Transformer
复制标题
DOI：
10.1587/transinf.2022mui0001
复制
发表时间：
2023-01-01
影响因子：
0.7
通讯作者：
Kinoshita, Yuma
中科院分区：
文献类型：
--
作者：
Kiya, Hitoshi;Iijima, Ryota;Kinoshita, Yuma
在本文中，我们提出了一种组合使用的变换图像和视觉Transformer（ViT）模型转换的秘密密钥。我们第一次表明，使用普通图像训练的模型可以直接转换为基于ViT架构使用加密图像训练的模型，并且转换后的模型的性能与使用密钥加密的测试图像时使用普通图像训练的模型相同。此外，该方案不需要任何专门准备的数据用于训练模型或网络修改，因此它也允许我们轻松地更新密钥。在一个实验中，所提出的计划的有效性进行评估的性能退化和模型保护性能在CIFAR-10数据集上的图像分类任务。
In this paper, we propose a combined use of transformed images and vision transformer (ViT) models transformed with a secret key. We show for the first time that models trained with plain images can be directly transformed to models trained with encrypted images on the basis of the ViT architecture, and the performance of the transformed models is the same as models trained with plain images when using test images encrypted with the key. In addition, the proposed scheme does not require any specially prepared data for training models or network modification, so it also allows us to easily update the secret key. In an experiment, the effectiveness of the proposed scheme is evaluated in terms of performance degradation and model protection performance in an image classification task on the CIFAR-10 dataset.