Shulou(Shulou.com)11/24 Report--

CTOnews.com, Oct. 17, Nvidia launched RTX VSR video super-resolution technology earlier this year, which requires the use of GeForce RTX 40 or 30 series GPU. Today, Nvidia announced that the technology is updated in version 1.5 to support RTX 20 series graphics cards.

According to reports, Nvidia RTX video super resolution version 1.5 can provide higher overall graphics fidelity, retain more details, support local video and support for GeForce RTX 20 series desktops and laptops GPU. RTX VSR's AI model has been retrained to more accurately identify the difference between fine details and compression artifacts, thus better preserving image details. RTX VSR 1.5 is available as part of the latest Game Ready driver and is now available for download.

The introduction of RTX VSR video super resolution technology attached to CTOnews.com is as follows:

RTX VSR is a breakthrough in AI image processing. It goes beyond the traditional edge detection and feature sharpening technology, and greatly improves the quality of live video content.

RTX VSR can reduce or eliminate artifacts caused by compressed video, such as blocking effect, ringing artifacts on edges, disappearance of high-frequency details and banding in flat areas, while reducing texture loss. It also sharpens edges and enriches details. This technique uses deep learning network to enlarge the image and reduce the compression artifacts in a single reasoning. The network analyzes low-resolution video frames and predicts residual images at the target resolution. Then, the residual image is superimposed on the traditional image enhancement to correct the artifact error and sharpen the edge to match the output resolution.

The deep learning network uses different degrees of compressed content for training. It learns the types of compression artifacts that do not exist in uncompressed images but exist in low-resolution or low-quality video, and uses these artifacts and their uncompressed images as a reference for network training. The Deep Learning Network also conducts extensive visual assessments to ensure that the generated models are suitable for almost all real-world and game content.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.