A Privacy Preserving Method with a Random Orthogonal Matrix for ConvMixer Models
A Privacy Preserving Method with a Random Orthogonal Matrix for ConvMixer Models
复制标题
DOI：
10.48550/arxiv.2301.03843
复制
发表时间：
2023-01
期刊：
影响因子：
--
通讯作者：
Rei Aso;Tatsuya Chuman;H. Kiya
中科院分区：
文献类型：
--
作者：
Rei Aso;Tatsuya Chuman;H. Kiya
提出了一种基于ConvMixer模型的隐私保护图像分类方法。为了保护测试图像的视觉信息，将测试图像分成块，然后使用随机正交矩阵对每个块进行加密。此外，在ConvMixer的嵌入结构的基础上，用用于加密测试图像的随机正交矩阵对用普通图像训练的ConvMixer模型进行变换。所提出的方法使我们不仅可以使用相同的分类精度的ConvMixer模型，而不考虑隐私保护，但也提高了对各种攻击的鲁棒性相比，传统的隐私保护学习。
In this paper, a privacy preserving image classification method is proposed under the use of ConvMixer models. To protect the visual information of test images, a test image is divided into blocks, and then every block is encrypted by using a random orthogonal matrix. Moreover, a ConvMixer model trained with plain images is transformed by the random orthogonal matrix used for encrypting test images, on the basis of the embedding structure of ConvMixer. The proposed method allows us not only to use the same classification accuracy as that of ConvMixer models without considering privacy protection but to also enhance robustness against various attacks compared to conventional privacy-preserving learning.