We introduce a real-time deep learning-based face synthesis technology for photoreal AI-avatars, and demonstrate two novel applications. We showcase the first zero-shot real-time deepfake system allowing anyone to swap their faces with another subject. Then, we demonstrate how this technology can enable an AI-based photoreal virtual assistant.
Typically, TV producers of motor-racing programs manually overlay visuals to provide on-screen context, such as a driver's name, position, or photo. Chroma Tools, deployed on live Formula E broadcast, automates these tasks and enables dynamic overlays that track the racers as they appear on the screen in strict real time.
DrawmaticAR is an AI-AR app capable of creating interactive 3D animated experiences from story words written in the designated section on an AR marker (Magic Paper).
How can VR/AR platforms personalize the user experience? I discuss how Looxid Link, the most VR-compatible, brainsensing technology, can connect users' minds to VR by visualizing, interacting with, and analyzing the users' minds in VR by demonstrating realtime examples using users' EEG feature and mind indexes.
Our tool allows artists to create living paintings or stylize a live video stream using their own artwork with minimal effort. While an artist is painting the image, our framework learns their artistic style on the fly and transfers it to the provided live video stream in real time.
New ObjectLandscapeTerrain system in UNIGINE Engine fulfils a very challenging combination of requirements: very dense details (down to 1 mm per pixel) together with huge terrain size (up to 10,000x10,000 km), real-time terrain geometry modification, nondestructive team collaboration, and binoculars/scope support (up to x20 zoom).
Sketch-to-Art is an AI tool that allows creatives to sketch an idea and get fully rendered images, stylized the way they want in real time. Users can define a style by either choosing a reference image, or a group of images, and selecting an artist, or an art movement.
Demonstrate the technological innovations behind the delivery of Millennium Falcon: Smugglers Run interactive attraction. This will show and explain the pieces of technology we had to write to achieve our goals of high fidelity, high resolution and high frame rate for the experience.
Existing volumetric capture systems require many cameras and lengthy post processing. We introduce the first system that can capture a completely clothed human body (including the back) using a single RGB webcam and in real time. Our deep-learning-based approach enables new possibilities for low-cost and consumeraccessible immersive teleportation.