In latest instances, display screen image high quality has transcended excessive definition – nowadays, it’s all about ultra-high definition. Although right now’s {hardware} has been developed to assist not simply 4K content material however 8K content material as nicely, solely lately has content material begun to really be produced to 8K high quality requirements. Therefore, with the intention to enhance the video show high quality of content material made earlier than 8K existed, upscaling know-how exists to maintain up with right now’s ultra-high definition customary.
Samsung Electronics’ 2020 QLED 8K TV lineup constantly receives reward for its units’ revolutionary show high quality, sound, design and lots of different areas. In explicit, the vary showcases a formidable evolution on the subject of its video upscaling know-how, on the core of which is Samsung’s AI Quantum Processor 8K.
Samsung Newsroom interviewed the crew from the Picture Quality Solution Lab of Samsung Electronics’ Visual Display Business to study extra about this revolutionary know-how.
(From left) Hyungjun Lim, Hyunseung Lee, Cheon Lee and Seungho Park of the Picture Quality Solution Lab of Samsung Electronics’ Visual Display Business
Leveling Up Deep Learning for Optimized Upscaling
When a low-resolution video performs on a high-resolution show, the ensuing picture high quality may be beneath a viewer’s expectations. AI (synthetic intelligence) upscaling is a know-how that converts low-resolution movies into ones of extra vivid and pure resolutions by analyzing the video information and adapting it intelligently.
In 2018, Samsung Electronics applied its machine learning-based AI 8K upscaling know-how for the primary time. This know-how harnesses machine studying whereby a human-designed algorithm intelligently analyzes the varied video qualities, together with line, mild and coloration, and improves every of those for an improved show decision. Such deep studying capabilities allow a TV to research and course of video options while concurrently studying because of a synthetic neural community that acts in the identical approach because the human mind.
The members of Samsung’s Picture Quality Solution Lab targeted on growing deep studying algorithms that allow shows to at all times showcase optimum video qualities with out human intervention. “Deep learning enables more precise and efficient image quality improvements than can be achieved by humans alone,” famous Hyungjun Lim, who developed the AI upscaling know-how’s texture era technique.
This operational benefit is what led the crew to develop an AI upscaling know-how that mixes machine studying with deep studying – Samsung’s AI Quantum Processor 8K. “Previous machine learning technology brought about enhanced sharpness to display picture quality, but now our technology can also offer more delicate texturing,” famous Hyunseung Lee, who developed enhanced edge sharpness for the AI Quantum Processor 8K. “Images with complicated textures, like mountain or grass landscapes, can now be upscaled into 8K quality more naturally.”
Solving the Conundrum of Deep Learning
Although deep studying has limitless potential, there are nonetheless obstacles to be overcome. The Picture Quality Solution Lab crew needed to undergo numerous trial-and-error situations with the intention to make their know-how market-ready. “It was difficult for us, at first, to clearly follow and comprehend the algorithm being developed by the artificial neural network of the processor,” highlighted Cheon Lee, the developer of the AI Quantum Processor 8K’s community construction and studying technique. “The artificial neural network’s high hardware chip power consumption was an issue we had to solve, too.”
After a prolonged improvement course of, Samsung was ultimately in a position to resolve the issues that had arisen because of its vary of differentiated applied sciences. “We achieved the extremely environment friendly show high quality efficiency of our…
