Near Optimal Per-Clip Lagrangian Multiplier Prediction in HEVC

TL;DR

This paper introduces a neural network-based method to predict optimal Lagrangian multipliers for HEVC video encoding, significantly reducing computational load while maintaining high bitrate savings.

Contribution

The work presents a novel neural network approach for per-clip Lagrangian multiplier prediction that achieves near-optimal bitrate savings with less computational effort.

Findings

Achieves BD-Rate improvement in about 90% of tested videos.

Maintains comparable results to previous optimization methods.

Reduces computational complexity of per-clip parameter tuning.

Abstract

The majority of internet traffic is video content. This drives the demand for video compression to deliver high quality video at low target bitrates. Optimising the parameters of a video codec for a specific video clip (per-clip optimisation) has been shown to yield significant bitrate savings. In previous work we have shown that per-clip optimisation of the Lagrangian multiplier leads to up to 24% BD-Rate improvement. A key component of these algorithms is modeling the R-D characteristic across the appropriate bitrate range. This is computationally heavy as it usually involves repeated video encodes of the high resolution material at different parameter settings. This work focuses on reducing this computational load by deploying a NN operating on lower bandwidth features. Our system achieves BD-Rate improvement in approximately 90% of a large corpus with comparable results to previous work in direct optimisation.