MMSep 29, 2020

Performance of AV1 Real-Time Mode

arXiv:2009.14165v21 citations
Originality Synthesis-oriented
AI Analysis

This work addresses the need for better real-time codec performance evaluation, particularly relevant for digital interactions during COVID-19, but it is incremental as it focuses on testing existing codecs rather than proposing new methods.

The authors tackled the lack of literature on real-time codecs by introducing a new test setup with a paced reader to simulate real-time media capture conditions, comparing performance of H.264, H.265, VP8, VP9, and AV1 codecs on HD and full HD video sequences at 25 and 50 fps, with results including quality measurements using VMAF and multiple speed metrics.

With COVID-19, the interest for digital interactions has raised, putting in turn real-time (or low-latency) codecs into a new light. Most of the codec research has been traditionally focusing on coding efficiency, while very little literature exist on real-time codecs. It is shown how the speed at which content is made available impacts both latency and throughput. The authors introduce a new test set up, integrating a paced reader, which allows to run codec in the same condition as real-time media capture. Quality measurements using VMAF, as well as multiple speed measurements are made on encoding of HD and full HD video sequences, both at 25 fps and 50 fps to compare the respective performances of several implementations of the H.264, H.265, VP8, VP9 and AV1 codecs.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes