Privacy-Preserving ConvMixer Without Any Accuracy Degradation Using Compressible Encrypted Images

We propose an enhanced privacy-preserving method for image classification using ConvMixer, which is an extremely simple model that is similar in spirit to the Vision Transformer (ViT). Most privacy-preserving methods using encrypted images cause the performance of models to degrade due to the influe...

Full description

Bibliographic Details

Published in:	Information
Main Authors:	Haiwei Lin, Shoko Imaizumi, Hitoshi Kiya
Format:	Article
Language:	English
Published:	MDPI AG 2024-11-01
Subjects:	privacy preserving image classification patch embedding image encryption access control
Online Access:	https://www.mdpi.com/2078-2489/15/11/723