LG: Viewing With Depth

15 Aug 2019

Deep learning is now being applied to how the human eye processes information

You don’t have to be actually old to be old enough to remember when the concept of self-driving cars was only the stuff of sci-fi movies. And then suddenly, we’re at that point in history when we have safe, efficient, autonomous vehicles in real-world streets, going at speeds that human-driven cars go.

Technologies that bring us such wonders are made possible by the availability of massive amounts of data—millions of collected images and thousands of hours of video from which to extract information—and what is called deep learning.

Deep learning is a machine learning method that employs artificial neural networks to automate the process of analyzing images and detecting objects. Deep learning is widely used where big data is available, from healthcare to electronics to social media to defense, and everything in between.

LG Electronics has tapped into the exciting progress that has been made by such studies, and used the results to innovate its latest TV lineup. It recently rolled out the industry-leading OLED TVs (the LG Signature OLED TV W9, LG OLED TV E9, and LG OLED TV C9) and the NanoCell TVs with ThinQ AI (the SM94, SM90, SM86 and SM81). These lines of products boast LG’s second-generation α9 and α7 intelligent processors and deep-learning algorithm.

The α9 Gen 2 processor is able to understand how the human eye perceives images in different lighting, and can finely adjust the tone mapping curve in accordance to ambient conditions to achieve the optimal level of screen brightness. It uses the TV’s ambient light sensor to measure external light, automatically changing the screen settings to compensate as required. Its advanced AI then refines High Dynamic Range content. In brightly lit settings, it can transform dark, shadow-filled scenes into easily discernible images, without sacrificing depth or making colors seem unnatural or oversaturated.

With this cutting-edge innovation in tailoring the user’s television experience to be more intuitive, fluid, and human, deep learning has come full circle in that it initially started by processing large volumes of images to detect data with an accuracy surpassing human performance. And now deep learning is being used to understand how the human eye processes information, and using that understanding to deliver products that take TV viewing to the next level.