返回
A Layer-Wise Extreme Network Compression for Super Resolution
DOI:10.1109/ACCESS.2021.3090404.png)
摘要
En 中文
Deep neural networks (DNNs) for single image super-resolution (SISR) tend to have large model size and high computation complexity to achieve promising restoration performance. Unlike image classification, model compression for SISR has rarely been studied. In this paper, we found out that DNNs for image classification and SISR have often different characteristics in terms of layer importance. That is, contrary to the DNNs for image classification, the performance of SISR networks hardly decrease even if a few layers are eliminated during inference. This is due to the fact that they typically consist of a bunch of hierarchical and complex residual connections. Based on that key observation, we propose a layer-wise extreme network compression method for SISR. The proposed method consists of: i) reinforcement learning based joint framework for layer-wise quantization and pruning both of which are effectively incorporated into the search space; ii) a progressive preserve ratio scheduling that reflects importance in each layer more effectively, yielding much higher compression efficiency. Our comprehensive experiments show that the proposed method can effectively be applied to the existing SISR networks, thus extremely reducing the model size up to 97% (i.e., 1 bit per weight on average) with marginal performance degradation compared to the corresponding full-precision models.
Keyword:
Quantization (signal)
Image coding
Image reconstruction
Adaptation models
Degradation
Convolution
Task analysis
Single image super resolution
model compression
layer-wise
quantization
pruning
joint learning
reinforcement learning
AI总结
对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
3.6
论文数:
9.8W
被引数:
29.4W
机构
引用论文
Under arrest: cytostatic factor (CSF)-mediated metaphase arrest in vertebrate eggs被捕: 脊椎动物卵中细胞生长抑制因子 (CSF) 介导的中期停滞
Calcium-modulating cyclophilin ligand regulates membrane trafficking of postsynaptic GABAA receptors
Red, green, and blue electrochromism in ambipolar poly(amine–amide–imide)s based on electroactive tetraphenyl‐p‐phenylenediamine units基于电活性四苯基 p-苯二胺单元的双极性聚 (胺-酰胺-酰亚胺) 中的红色,绿色和蓝色电致变色

