Privacy-Preserving Semantic Segmentation Using Vision Transformer

In this paper, we propose a privacy-preserving semantic segmentation method that uses encrypted images and models with the vision transformer (ViT), called the segmentation transformer (SETR). The combined use of encrypted images and SETR allows us not only to apply images without sensitive visual i...

Бүрэн тодорхойлолт

Номзүйн дэлгэрэнгүй

Үндсэн зохиолчид:	Hitoshi Kiya, Teru Nagamori, Shoko Imaizumi, Sayaka Shiota
Формат:	Artigo
Хэл сонгох:	Inglês
Хэвлэсэн:	MDPI AG 2022-08-01
Цуврал:	Journal of Imaging
Нөхцлүүд:	semantic segmentation vision transformer segmentation transformer privacy-preserving
Онлайн хандалт:	https://www.mdpi.com/2313-433X/8/9/233
Шошгууд:	Шошго нэмэх Шошго байхгүй, Энэхүү баримтыг шошголох эхний хүн болох!