ProxIQA: A Proxy Approach to Perceptual Optimization of Learned Image Compression

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, these simple norms are not very consistent with human perception. Here, we describe a different "proximal" approach to optimize image analysis networks against quantitative perceptual models. Specifically, we construct a proxy network, broadly termed ProxIQA, which mimics the perceptual model while serving as a loss layer of the network. We experimentally demonstrate how this optimization framework can be applied to train an end-to-end optimized image compression network. By building on top of an existing deep image compression model, we are able to demonstrate a bitrate reduction of as much as $31\%$ over MSE optimization, given a specified perceptual quality (VMAF) level.

12 pages, 12 figures, 5 tables

References in corpus (7)

• Variational image compression with a scale hyperprior
• End-to-end Optimized Image Compression
• Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment
• The Perception-Distortion Tradeoff
• Adaptive Debanding Filter
• Speeding up VP9 Intra Encoder with Hierarchical Deep Learning Based Partition Prediction
• GAN- vs. JPEG2000 Image Compression for Distributed Automotive Perception: Higher Peak SNR Does Not Mean Better Semantic Segmentation

Cited by in corpus (4)

• UGC-VQA: Benchmarking Blind Video Quality Assessment for User Generated Content
• RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content
• RankDVQA: Deep VQA based on Ranking-inspired Hybrid Training
• Scale-Consistent Fusion: from Heterogeneous Local Sampling to Global Immersive Rendering