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...

Cijeli opis

Bibliografski detalji

Glavni autori:	Haiwei Lin, Shoko Imaizumi, Hitoshi Kiya
Format:	Artigo
Jezik:	Inglês
Izdano:	MDPI AG 2024-11-01
Serija:	Information
Teme:	privacy preserving image classification patch embedding image encryption access control
Online pristup:	https://www.mdpi.com/2078-2489/15/11/723
Oznake:	Dodaj oznaku Bez oznaka, Budi prvi tko označuje ovaj zapis!