Speed-Distortion Optimization: Tradeoffs in Open-Source HEVC Encoding

The high-efficiency video coding (HEVC) video compression standard, which drives a significant increase in coding efficiency, is an essential technology for delivering ultrahigh-definition content in the high dynamic range (HDR). This improved coding efficiency comes at a significant cost in computational complexity. Adaptive streaming and content-adaptive encoding significantly increase the compute required for encoding for the additional encodes and analysis required to support the corresponding use cases. All these factors combined require a careful analysis of tradeoffs to avoid exorbitant infrastructure costs of transition to HEVC. This paper discusses how the throughput of an HEVC encoder can be optimized to enable a cost-efficient distributed encoding system running on commodity hardware. We chose the popular open-source ×265 video encoder for this study due to its wide adoption. We first examine the ×265 presets, which combine various performance and encoding efficiency tradeoffs into a single operational point. We then detail some additional options that enable additional analysis on a selective subset of cases, such as per frame or per coding tree unit; these options are representative of several such options that are not exposed via the presets of ×265. Specifically, we discuss options pertaining to selective in-loop filtering and dynamically retargeting rate-distortion optimization (RDO) tools depending on the quality fluctuations encountered at runtime. We present tradeoff results for real-world 4K HDR content to demonstrate the applicability of these coding tools to real content used for over-the-top (OTT) and live streaming
.