The Facebook research division proposed a new approach to hand-tracking.
Facebook showed a system for real-time hand-tracking that is meant for virtual and augmented reality experiences.
First, check out an overview by the Two-Minute Papers channel:
"Using four fisheye monochrome cameras, our system generates accurate and low-jitter 3D hand motion across a large working volume for a diverse set of users," wrote the team. "We achieve this by proposing neural network architectures for detecting hands and estimating hand keypoint locations."
The research division states that they designed scalable, semi-automated mechanisms to collect a large and diverse set of ground truth data using a mix of manual annotation and automated tracking. They also used a detection-by-tracking method to boost smoothness while reducing the computational cost (the optimized system runs at 60Hz on PC and 30Hz on a mobile processor).
You can find the full paper here. Don't forget to join our new Telegram channel, our Discord, follow us on Instagram and Twitter, where we are sharing breakdowns, latest news, awesome artworks, and more.