Structured Pruning and Quantization for Learned Image Compression

TL;DR

This paper introduces a structured pruning and quantization method for Learned Image Compression models, significantly reducing computational costs while preserving compression performance, and demonstrating the effectiveness of combined model compression techniques.

Contribution

The paper proposes a novel NAS-based structured pruning approach for LIC models that maintains performance and can be combined with quantization for further compression.

Findings

Model size reduced without BD-Rate performance drop

Pruning integrated with quantization achieves additional compression

Source code made publicly available

Abstract

The high computational costs associated with large deep learning models significantly hinder their practical deployment. Model pruning has been widely explored in deep learning literature to reduce their computational burden, but its application has been largely limited to computer vision tasks such as image classification and object detection. In this work, we propose a structured pruning method targeted for Learned Image Compression (LIC) models that aims to reduce the computational costs associated with image compression while maintaining the rate-distortion performance. We employ a Neural Architecture Search (NAS) method based on the rate-distortion loss for computing the pruning ratio for each layer of the network. We compare our pruned model with the uncompressed LIC Model with same network architecture and show that it can achieve model size reduction without any BD-Rate performance drop. We further show that our pruning method can be integrated with model quantization to achieve further model compression while maintaining similar BD-Rate performance. We have made the source code available at gitlab.com/viper-purdue/lic-pruning.