BriefGPT.xyz
Oct, 2019
ProxIQA：一种基于代理的感知优化方法用于学习图像压缩
ProxIQA: A Proxy Approach to Perceptual Optimization of Learned Image Compression
HTML
PDF
Li-Heng Chen, Christos G. Bampis, Zhi Li, Andrey Norkin, Alan C. Bovik
TL;DR
本文介绍一种基于感知模型的优化框架，通过构建一个代理网络(ProxIQA)，可以优化图像分析网络及图像压缩网络，从而在特定感知质量下，降低比特率的同时维持图像质量。
Abstract
The use of $\ell_p$ $(p=1,2)$ norms has largely dominated the measurement of loss in
neural networks
due to their simplicity and analytical properties. However, when used to assess the loss of
visual information
,
→