English

ERNet Family: Hardware-Oriented CNN Models for Computational Imaging Using Block-Based Inference

Machine Learning 2020-01-31 v2 Image and Video Processing Machine Learning

Abstract

Convolutional neural networks (CNNs) demand huge DRAM bandwidth for computational imaging tasks, and block-based processing has recently been applied to greatly reduce the bandwidth. However, the induced additional computation for feature recomputing or the large SRAM for feature reusing will degrade the performance or even forbid the usage of state-of-the-art models. In this paper, we address these issues by considering the overheads and hardware constraints in advance when constructing CNNs. We investigate a novel model family---ERNet---which includes temporary layer expansion as another means for increasing model capacity. We analyze three ERNet variants in terms of hardware requirement and introduce a hardware-aware model optimization procedure. Evaluations on Full HD and 4K UHD applications will be given to show the effectiveness in terms of image quality, pixel throughput, and SRAM usage. The results also show that, for block-based inference, ERNet can outperform the state-of-the-art FFDNet and EDSR-baseline models for image denoising and super-resolution respectively.

Keywords

Cite

@article{arxiv.1910.05787,
  title  = {ERNet Family: Hardware-Oriented CNN Models for Computational Imaging Using Block-Based Inference},
  author = {Chao-Tsung Huang},
  journal= {arXiv preprint arXiv:1910.05787},
  year   = {2020}
}

Comments

5 pages; appearing in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020

R2 v1 2026-06-23T11:42:20.333Z