High-throughput systolic array-based accelerator for hybrid transformer-CNN networks

In this era of Transformers enjoying remarkable success, Convolutional Neural Networks (CNNs) remain highly relevant and useful. Indeed, hybrid Transformer-CNN network architectures, which combine the benefits of both approaches, have achieved impressive results. Vision Transformer (ViT) is a signif...

Полное описание

Библиографические подробности

Главные авторы:	Qingzeng Song, Yao Dai, Hao Lu, Guanghao Jin
Формат:	Artigo
Язык:	Inglês
Опубликовано:	Springer 2024-10-01
Серии:	Journal of King Saud University: Computer and Information Sciences
Предметы:	Hardware accelerator Hybrid transformer-CNN Systolic array FPGA
Online-ссылка:	http://www.sciencedirect.com/science/article/pii/S1319157824002830
Метки:	Добавить метку Нет меток, Требуется 1-ая метка записи!