Share Your Science: Creating Volumetric Video of Real People in Virtual Reality

Experience real people on any virtual reality headset.

Joel Pitt, a machine learning researcher at 8i, shares how the startup is using NVIDIA GPUs and CUDA to put real volumetric video of humans in virtual reality environments.

8i’s technology allows creators to bring real people into volumetric virtual experiences by transforming HD video from multiple cameras into a fully volumetric recording of a human that viewers can walk around in virtual reality, augmented reality, and web.

Pitt mentions they are currently looking to leverage deep learning and cuDNN to improve the real-time reconstruction quality, such as the coloring and depth estimation to help make the most realistic humans as possible.



To learn more about volumetric video, watch an interview with 8i CEO Linc Gasking about their documentary #100Humans that was unveiled at the Sundance Film Festival early this year.

Share your GPU-accelerated science with us at http://nvda.ly/Vpjxr and with the world on #ShareYourScience.

Watch more scientists and researchers share how accelerated computing is benefiting their work at http://nvda.ly/X7WpH

 

About Brad Nemire

Brad Nemire leads the Developer Communications team at NVIDIA focused on evangelizing amazing GPU-accelerated applications. Prior to NVIDIA, he worked at Arm on the Developer Relations team. Brad graduated from San Diego State University.