# Architecture-aware Network Pruning for Vision Quality Applications

## TL;DR

This paper introduces an iterative, architecture-aware pruning method for CNNs that reduces computational costs in vision quality tasks like low-light imaging and super-resolution without sacrificing quality.

## Contribution

It presents a novel adaptive pruning algorithm that improves efficiency while maintaining image quality in high-resolution vision applications.

## Key findings

- Reduced MAC by 58% in low-light imaging
- Decreased MAC by 37% in super-resolution
- Lowered memory bandwidth by 20-40%

## Abstract

Convolutional neural network (CNN) delivers impressive achievements in computer vision and machine learning field. However, CNN incurs high computational complexity, especially for vision quality applications because of large image resolution. In this paper, we propose an iterative architecture-aware pruning algorithm with adaptive magnitude threshold while cooperating with quality-metric measurement simultaneously. We show the performance improvement applied on vision quality applications and provide comprehensive analysis with flexible pruning configuration. With the proposed method, the Multiply-Accumulate (MAC) of state-of-the-art low-light imaging (SID) and super-resolution (EDSR) are reduced by 58% and 37% without quality drop, respectively. The memory bandwidth (BW) requirements of convolutional layer can be also reduced by 20% to 40%.