RTX Video Super Resolution Firefox
This article is a part of the AI Decoded series, which shows off new RTX PC hardware, software, tools, and accelerations while demystifying AI by making the technology more approachable.
The well-known open-source browser Mozilla Firefox is the most recent partner to integrate NVIDIA RTX Video, a feature that use AI to enhance the quality of video on Windows workstations and PCs. The most recent version of the browser makes use of nearby NVIDIA RTX GPUs to improve video and streaming.
NVIDIA RTX Video
Pixel-Excellent
NVIDIA RTX Video, which was first unveiled at CES in January 2023, is an assortment of artificial intelligence (AI) video advancements designed to raise the calibre of videos viewed on browsers via services like YouTube, Amazon Video, and Disney+. The technology enhances the sharpness and detail of videos streamed on PCs and workstations with NVIDIA GeForce RTX graphics without the need for a higher-resolution source.
NVIDIA RTX Video consists of two components. Low-resolution video is upsized by NVIDIA RTX Video Super Resolution to produce clearer, more vivid images. In order to forecast what the higher-resolution version should look like, deep learning is used to analyse the lower-resolution footage. In order to minimise or completely remove compression artefacts and improve the output’s sharpness, the algorithm then mixes the anticipated image with a conventionally upscaled version.
NVIDIA RTX Video Super Resolution
Beyond edge recognition and feature sharpening, RTX Video Super Resolution‘s innovation in AI pixel processing significantly raises the quality of streamed video content.
In streaming video, blocky compression artefacts are a recurring problem. Issues with traditional upscaling often get worse when it comes to streaming material, making the visual experience less enjoyable for viewers. These problems might arise from the content itself, the server, or the client.
With RTX Video Super Resolution, lost texture is reduced and compression-related artefacts such blockiness, ringing artefacts around edges, washout of high-frequency features, and banding on flat areas are reduced or eliminated. It also brings details and edges to life.
The system makes use of a deep learning network that can complete compression artefact reduction and upscaling in a single pass. The residual image at the desired resolution is predicted by the network based on its analysis of the lower-resolution video frame. The remaining image is then sharpened to match the output resolution and artefact errors are corrected by superimposing it over a conventionally upscaled image.
Various compression settings and a broad variety of information are used to train the deep learning network. As a reference for network training, it learns about kinds of compression artefacts found in low-resolution or low-quality videos that are otherwise missing in uncompressed images. To guarantee that the resulting model works on almost all real-world and gaming content, extensive visual inspection is used.
NVIDIA RTX Video HDR
To enhance clarity, details, and vibrancy, NVIDIA RTX Video HDR goes one step further by using artificial intelligence (ANN) neural networks to analyse standard dynamic range (SDR) video content and add high-dynamic range (HDR) information.
NVIDIA RTX Video HDR enables producers and gamers to optimise the dynamic colour reproduction of their HDR panels while retaining fine details that could be accidentally lost as a result of video compression. This is achieved by using the power of Tensor Cores on GeForce RTX GPUs.
One of the RTX-powered applications that improves regular PC use, productivity, creative, and gaming is NVIDIA RTX Video HDR. NVIDIA Canvas creates realistic landscape visuals with simple brushstrokes, NVIDIA Broadcast powers up microphones and cameras, and NVIDIA Omniverse effortlessly integrates 3D apps and creative workflows. Discover premium Studio resources, such as the free NVIDIA Studio Drivers for RTX graphics card owners, which support the most recent updates to creative apps, AI-powered capabilities, and more.
Enabling NVIDIA RTX Video is equivalent to pressing the “remaster” button on the majority of the content consumers watch on a daily basis, given 90% of online video is 1080p or lower and SDR.
Adorable Foxy
With its most recent stable version (v126), Mozilla Firefox now supports NVIDIA RTX Video Super Resolution and HDR. Users can now enjoy AI-enhanced HDR, de-artifacting, and upscaling effects for web videos more easily than ever before.
Bobby Holley, chief technical officer of Firefox at Mozilla, stated, “Video is a core pillar of the modern web, and NVIDIA are committed to delivering a great experience for their users.” “For users with compatible GPUs, Mozilla is integrating NVIDIA RTX Video into Firefox to improve video quality.”
Firefox adds support for NVIDIA RTX Video to the list of Chromium-based browsers, which also includes Microsoft Edge and Google Chrome. Well-known video players like VLC also support RTX Video Super Resolution.
NVIDIA Studio platform
The NVIDIA Studio platform, which powers enthusiasts, aspiring artists, and pros in the field, enhances your creative process with RTX and AI. When combined with NVIDIA-only driver technology, industry-leading RTX GPUs from NVIDIA improve your creative apps with performance and capabilities that are truly remarkable. With NVIDIA Studio speeding up your ideas and workflows, you can accomplish your most ambitious projects to date.
NVIDIA RTX Video is simple to enable:
- Update to the most recent NVIDIA Studio, NVIDIA RTX Enterprise, or GeForce RTX Game Ready driver.
- Go to System > Display > HDR to make sure Windows HDR features are activated.
- To get RTX Video Enhancement, open the NVIDIA Control Panel and go to Adjust Video Image Settings.
- Turn on “High Dynamic Range” and “Super Resolution.”
Be aware that in order to use RTX Video HDR, a monitor or TV that supports HDR10 must be connected to an NVIDIA GeForce RTX or RTX professional GPU.
