Streaming high-resolution video typically comes with an inevitable trade-off between available bandwidth and quality of experience for the end user. Delivering uncompromising video quality typically requires excessively high bitrates, which can result in slow starts, video buffering and high content delivery network (CDN) and storage costs. As the percentage of IP traffic attributed to video increases (estimated to already surpass 82%), these problems are only exacerbated, driving greater urgency for new innovations to address these challenges.
Traditional solutions that attempt to minimize bandwidth without compromising quality are centered around the development of more intelligent video encoders; either by replacing rate control, quantization and prediction strategies within them, or the entirety of a standard video coding pipeline. The latter, however, is a particularly risky proposition for video encoding services, since it requires the creation of bespoke transport mechanisms and decoders across multiple client device types. Likewise, improvements generated by a standards-based codec remain severely constrained by its inherent compliance needs.