The Facebook AI Camera Team wrote on the company’s research blog today that bodies are much more difficult to track than faces.
Facebook solved this problem by retraining its own computer vision technology, which it already uses to fight revenge porn and to recommend friends to tag in uploaded pictures. The company taught a neural network to recognize different body poses and to distinguish between a person and the background behind them, allowing AR animations to be anchored to the proper body parts much like an animation rig in 3D animation.
The company’s AI research team boasted that the program takes up “only a few megabytes” of storage space and can run in real time on modern smartphones. This is good news for AR developers who don’t want to limit their apps to high-powered devices with external computing devices, such as the recently unveiled Magic Leap.
See the full story here: https://siliconangle.com/blog/2018/01/25/facebook-built-ai-enables-full-body-augmented-reality-experiences/