Youngo Park (left) and Kwangpyo Choi, from Samsung Research’s Visual Technology workforce
It’s official: We’ve entered the period of 8K TVs. Around the world, gross sales of 8K shows are steadily growing, and with TV producers consistently including extra choices to the combo, the 8K market is anticipated to proceed to develop.
There are, nonetheless, a couple of challenges that must be addressed earlier than viewers all over the world will be capable to get pleasure from 8K’s gorgeous visuals of their entirety. First, extra 8K content material will must be produced, and second, community connections must be made able to supporting the seamless streaming of 8K films and exhibits.
To tackle these points, researchers from Samsung Research, a sophisticated R&D hub inside Samsung Electronics’ SET Business, have developed an AI Codec generally known as AI ScaleNet. First launched at SDC19, this innovation permits the supply of 8K content material on networks that sometimes assist solely 4K speeds – with out the necessity for added infrastructure.
To be taught extra about AI ScaleNet’s improvement course of and potential functions, Samsung Newsroom interviewed Kwangpyo Choi and Youngo Park – researchers from Samsung Research’s Visual Technology workforce, and the primary people to recommend and develop one of these progressive know-how.
Supporting More 8K Content
AI ScaleNet makes use of deep studying know-how to reduce knowledge loss throughout compression and allow 8K content material to be streamed on networks with decrease bandwidth capabilities.
Here’s the way it works, in a nutshell: 8K content material is compressed to 4K high quality utilizing an AI downscaler and transmitted to the consumer’s TV, which makes use of AI to upscale the content material again to 8K high quality. When requested to explain the inspiration for the know-how, Choi specified a necessity for a brand new video compression and transmission know-how with a purpose to accommodate the ever-evolving media panorama, and to deal with associated technical challenges.
“Multimedia content is rapidly moving online, so figuring out how to overcome bandwidth limitations for UHD content transmission has become a major task,” stated Choi.
“With AI ScaleNet, users will be able to enjoy 8K-quality content, even at lower bandwidths,” added Park. “In fact, during periods of high network traffic, when speeds tend to drop, viewers can still expect content to be presented at a relatively higher quality.”
AI ScaleNet addresses bandwidth limitations inside trendy community infrastructures with what’s generally known as an adaptive AI Codec. “AI Adaptive Bit Rate Streaming refers to AI technology that adapts to various bandwidths and adjusts the resolution to enable seamless streaming,” stated Choi.
The distinction the know-how makes is particularly clear when utilizing over-the-top (OTT) companies, the place viewers sometimes discover adjustments in image high quality. As Choi defined, nonetheless, Samsung’s AI Codec and AI ScaleNet know-how make the most of AI that adapts to adjustments in bandwidth and optimizes high quality, so customers benefit from the highest-quality streaming expertise attainable.
A Groundbreaking Innovation
As the researchers described, the event course of wasn’t with out its difficulties. “The most difficult part was ensuring compatibility with existing codec compression technologies,” stated Choi. “How to design an easy-to-use algorithm that would be compatible with existing technologies, while delivering the intended performance at the same time was the question.”
Choi additionally famous difficulties that derived from gaps in perspective between the analysis and improvement groups. There have been a number of issues associated to complexity, design and construction which researchers took without any consideration, and have been questioned by the builders.
“We have been in a position to obtain this end result due to the synergy of each our workforce members, who labored carefully to attain the identical purpose, and colleagues who have been targeted on the know-how’s…
