Substitutional Neural Image Compression

by   Xiao Wang, et al.

We describe Substitutional Neural Image Compression (SNIC), a general approach for enhancing any neural image compression model, that requires no data or additional tuning of the trained model. It boosts compression performance toward a flexible distortion metric and enables bit-rate control using a single model instance. The key idea is to replace the image to be compressed with a substitutional one that outperforms the original one in a desired way. Finding such a substitute is inherently difficult for conventional codecs, yet surprisingly favorable for neural compression models thanks to their fully differentiable structures. With gradients of a particular loss backpropogated to the input, a desired substitute can be efficiently crafted iteratively. We demonstrate the effectiveness of SNIC, when combined with various neural compression models and target metrics, in improving compression quality and performing bit-rate control measured by rate-distortion curves. Empirical results of control precision and generation speed are also discussed.


page 1

page 2

page 3

page 4

page 5

page 6

page 7

page 8


Estimating the Resize Parameter in End-to-end Learned Image Compression

We describe a search-free resizing framework that can further improve th...

Rate Distortion Characteristic Modeling for Neural Image Compression

End-to-end optimization capability offers neural image compression (NIC)...

Lossy Compression with Distortion Constrained Optimization

When training end-to-end learned models for lossy compression, one has t...

Revisiting Pre-analysis Information Based Rate Control in x265

Due to the excellent compression and high real-time performance, x265 is...

Variable-Bitrate Neural Compression via Bayesian Arithmetic Coding

Deep Bayesian latent variable models have enabled new approaches to both...

The Rate-Distortion-Accuracy Tradeoff: JPEG Case Study

Handling digital images is almost always accompanied by a lossy compress...

Pseudocylindrical Convolutions for Learned Omnidirectional Image Compression

Although equirectangular projection (ERP) is a convenient form to store ...