Differentiable bit-rate estimation for neural-based video codec enhancement

TL;DR

This paper introduces a differentiable bit-rate estimation method for neural-based video codec enhancement, enabling efficient gradient computation and improving neural network training for video compression.

Contribution

A novel mathematical model for bit-rate estimation that accounts for statistical dependencies, providing closed-form formulas for estimates and gradients, reducing computational complexity.

Findings

Accurately estimates HEVC/H.265 bit-rates

Efficiently incorporates statistical dependencies in estimation

Reduces computational complexity for gradient calculation

Abstract

Neural networks (NN) can improve standard video compression by pre- and post-processing the encoded video. For optimal NN training, the standard codec needs to be replaced with a codec proxy that can provide derivatives of estimated bit-rate and distortion, which are used for gradient back-propagation. Since entropy coding of standard codecs is designed to take into account non-linear dependencies between transform coefficients, bit-rates cannot be well approximated with simple per-coefficient estimators. This paper presents a new approach for bit-rate estimation that is similar to the type employed in training end-to-end neural codecs, and able to efficiently take into account those statistical dependencies. It is defined from a mathematical model that provides closed-form formulas for the estimates and their gradients, reducing the computational complexity. Experimental results demonstrate the method's accuracy in estimating HEVC/H.265 codec bit-rates.