Per-Title Encoding Optimization and Workflow
Learn more about encoding optimization at Streaming Media East 2020.
Read the complete transcript of this clip:
Steve Geiger: The goal here of per-title encoding is to optimize the bitrate ladder for every single encode, and then what that's going to do is allow us to really focus and try to fit into that relatively narrow band that the human perceptual system can see. So if you go above that, you're going to waste a lot of bits in your encodes. If you go below that, you're going to see a lot of artifacts that your users are going to pick up on. Ultimately, what this allows us to do is try to reduce the amount of cost and improve the amount of quality that we want to put into every encode.
So that's how it works. We're talking about bitrate reduction, which equates to saving money. We'll talk about storage reduction, which equates to saving money, but then there's also the quality improvement aspect here, which means that these are a better experience in every dynamic. Some things that people don't think about often when we talk about per-title encoding is that it provides a consistent perceptual quality across the media library. If I'm always optimizing for that perceptual band, then you get a really consistent amount of quality across your scenario. That also allows you to get better performance in low-bandwidth scenarios, beecause we're not including step sizes, and it gets a better quality to those users who are ready to maybe reduce infrastructures.
This is how our technology works. It's very addable. We do an overall complexity analysis of the file, taking samplings across the whole asset, and then we do what we call a convex hull prediction, which ultimately amounts to doing a lot of sampling goods and using an algorithm to pick where all of those different bitrates and resolutions perform best, and then trying to draw the smoothest curve that we can within that perceptual bend.
Something that we try and also give is not just the optimal set of encodes, but also, everybody has different business constraints, and so we try to give you a lot of flexibility, so setting upper bounds, lower bounds, step sizes, target quality levels, things like this that make it a little bit more business-friendly for people who have specific use cases.
Video encoding began as a one-dimensional data rate adjustment that reflected the simple reality that all videos encode differently is now a complex analysis that incorporates frame rate, resolution, color gamut, and dynamic ranges, as well as delivery network and device-related data, along with video quality metrics.
If you're looking to launch an OTT service, these are the questions you need to ask potential solutions providers.
With so many moving parts in the streaming workflow, are best practices even possible? Five industry leaders share their insights into how to make all of the pieces fit
Streaming Learning Center Principal Jan Ozer explains per-title encoding and rates the different per-title encoding technologies in this clip from his presentation at Streaming Media East 2019.
SSIMWAVE Chief Science Officer Zhou Wang identifies the key evolutionary points in the development of per-title encoding to greater efficiency in this clip from his presentation at Streaming Media East.
SSIMWAVE Chief Science Officer introduces the key benefits of per-title encoding optimization in this clip from his presentation at Streaming Media East.