Nvidia’s New RTX Broadcast Engine Features AI-Powered Green Scre

Nvidia’s latest RTX-series GPUs are the first to feature real-time ray tracing, which can make games look prettier than ever. However, they’re also powerful computing platforms for AI development. Nvidia has demoed a new AI application powered by the Tensor Cores in RTX cards called RTX Broadcast Engine. Among other things, the new Broadcast Engine can clip the background out of a video feed to create an artificial “green screen.”

RTX Greenscreen is similar to features integrated with select webcams from Logitech C922x and Razer Stargazer, but Nvidia’s solution should work with any video feed. Those webcams with custom background removal tend not to work very well anyway. If Nvidia’s solution works well, it could make those devices much less attractive.

The RTX Broadcast Engine also features a face-tracking augmented reality (AR) feature called RTX AR. It recognizes faces with enough detail to map your features and render them on a 3D model. Again, this feature is similar to what you can get with specific pieces of hardware like the iPhone. That device uses IR dots and special cameras to map your face and produce “Animoji” versions of your face. Nvidia doesn’t need real depth data — the GPU just figures it out with AI.

Finally, there are RTX Style Filters, which use AI to apply filters to a full video. The idea here is you can use the Broadcast Engine to match the style of a game or artwork. In the demo video, Nvidia applies a “Starry Night” filter to a video in real-time. We don’t know how well this will be able to match the look of a game, but it could make for some cool streaming effects.

Developers can apply for early access to the RTX Broadcast Engine right now on Nvidia’s site. It will roll out to everyone “in the coming months.” Nvidia is also working with the makers of popular streaming software OBS to integrate the RTX Greenscreen feature.

Continue reading

Introducing TRACBench, a New AI-Powered Transcoding Benchmark

Three of the fastest workstation CPUs on Earth slug it out in an AI-powered transcoding benchmark we designed to scale from dual-core to 64-core systems.

AI-Powered Electronic Design Automation Tools Could Redefine Chipmaking

Samsung has announced it's using AI EDA tools to build its Exynos processors. Expect more announcements in the future, but geared more towards gradual adoption than a sudden surge.

IBM Builds an AI-Powered Electronic Tongue

The tongue, called Hypertaste, can detect and analyze a liquid's chemical composition.

Google Is Testing Its Latest AI-Powered Robot In a Kitchen

Using a 540-billion parameter language network, this wheeled robot can assess its own capabilities, its environment, and the task at hand, then break that task into smaller sub-tasks to achieve the desired goal.