Accurate and Bandwidth Efficient Architecture for CNN-based Full-HD Super-Resolution
Journal
Proceedings - IEEE International Symposium on Circuits and Systems
Journal Volume
2018-May
Date Issued
2018
Author(s)
Abstract
CNN-based super-resolution methods achieve great performance but come with high bandwidth requirements. This paper proposes a bandwidth efficient CNN-based architecture for super resolution supporting up to Full HD images at 60 fps. The bandwidth efficiency is achieved by layer fusion with data reuse scheduling and dynamic quantization. To maintain a high utilization of processing elements, the deconvolution operation is decomposed into a several convolution operations. The proposed architecture provides a 94.8 % bandwidth reduction. Compared to the software implementation, less than 1 % performance drop is induced in this VLSI implementation. The enables the proposed architecture to obtain state-of-the-art accuracy compared to existing super-resolution hardware implementations.
SDGs
Type
conference paper