Per Clip Lagrangian Multiplier Optimisation for HEVC

TL;DR

This paper explores optimizing the Lagrangian multiplier in HEVC to improve video compression efficiency, achieving significant bitrate savings through direct BD-Rate maximization on real-world YouTube videos.

Contribution

It introduces a method to directly optimize the Lagrangian multiplier in HEVC for individual clips, resulting in notable BD-Rate improvements over traditional approaches.

Findings

Up to 21% BD-Rate improvement for individual clips

Optimization using BD-Rate as the objective yields additional bitrate savings

Effective on a realistic YouTube video corpus

Abstract

The majority of internet traffic is video content. This drives the demand for video compression in order to deliver high quality video at low target bitrates. This paper investigates the impact of adjusting the rate distortion equation on compression performance. An constant of proportionality, k, is used to modify the Lagrange multiplier used in H.265 (HEVC). Direct optimisation methods are deployed to maximise BD-Rate improvement for a particular clip. This leads to up to 21% BD-Rate improvement for an individual clip. Furthermore we use a more realistic corpus of material provided by YouTube. The results show that direct optimisation using BD-rate as the objective function can lead to further gains in bitrate savings that are not available with previous approaches.