LG Display engineers together with specialists from South Korean University Sogang (Sogang University) presented a deep learning technology that reduces the delay between real motion and its reflection in VR, and also reduces image blur. It is these factors that make people nauseous when immersed in virtual reality. The joint development of South Korean scientists and specialists LG Display solves this problem.
Accuracy and plans
To mitigate the out of sync, developers often increase the image resolution, but this greatly loads the memory of the device. Instead, LG Display and Sogan University teams have developed an algorithm that reduces real-time blur and blur levels by 80%.
In addition, experts have agreed to work together on a tool to measure these indicators. It is based on data from a machine that simulates neck movement and an optical system that replicates the structure of the visual cortex.
The task of synchronizing actions with the image last year was decided by the Russian company VRT. She used the reverse kinematics algorithm, which calculates the possible movements of the user, based on data from sensors and information about the structure of the body.