A new patent from Sony regarding image reconstruction technology along the lines of NVIDIA’s Deep Learning Super Sampling technology that could be used for the upcoming PlayStation 5 console has recently surfaced online.
The new patent, which was filed in 2019, was alone published on July 23, details technology capable of reconstructing an image using a number of reference images through machine learning.
The patent is unfortunately only available in Japanese for now, so many details are still unknown, but from the general description it seems that this system will be an improvement over Checkerboard Rendering used by the PlayStation 4 Pro console .
An information processing apparatus for acquiring a number of reference images obtained by imaging an object to be reproduced, acquiring a number of converted images obtained by enlarging or reducing each of the reference images, performing machine learning using a number of images to are learned, as learning data, which includes the plurality of converted images, and generating pre-learned data used to generate a reproduction image representing the appearance of the object.
Since companies are filing patents for technology that will ultimately go unused, it is not certain that the PlayStation 5 will include this image reconstruction technology which is detailed in the new Sony patent. NVIDIA’s DLSS 2.0 technology proves how such technology can result in better visual quality at lower resolutions, so there’s no question that something similar would greatly benefit the next generation of console games.
The PlayStation 5 will be launched worldwide later this year. A precise release date, as well as final pricing information for both the standard edition and the digital edition, is yet to be confirmed. We’ll keep you posted on the console as soon as an announcement is made, so stay tuned for the latest news.