Multimodal Virtual Reality

UCLA, California | May 2020 - Aug 2021

W.M. Keck Centre for Neurophysics

I happened to come across this custom VR system at Dr. Mayank Mehta's lab at the W.M. Keck Centre for Neurophysics and I did not hesitate to go ahead and drop him an email expressing my interest in working on the custom VR system. It was a great blend of my interest in cognitive neuroscience and my proclivity for computer science and systems.

The custom VR System

The custom VR system at Dr. Mehta's lab allows for multimodal VR experiences for small animals on which non-invasive neuroscientific experiments can be conducted.
Virtual reality allows the experimenters to design experiments unencumbered by physical constraints. Often, the virtual experimental setups will be far from what we see in the real-world, which gives the experimenters an extra dimension of exploration.
The multimodal nature of the VR system is unlike anything else, allowing for a multitude of complex stimuli, all while being able to soundly carry out electrophysiological recordings.

My contribution

I was tasked with building the second generation of the VR system that had been built during 2009-2013. A lot has changed in the field of computing techniques since then and the system would work even better with an update.
I rebuilt the VR around a Unity graphics engine, replacing the earlier OGRE3D graphics engine. I also incorporated computer vision based fictive path tracking to replace the earlier optic mice based motion tracking. I am currently in the process of integrating a computer vision based eye-tracking system to replace the current LED head tracking solution, which is acting as a proxy for eye-tracking.
I am also coordinating the stimuli, the brain recording data with the motion data of the animal inside the virtual world to make the data readily available for experimenters.

Tools used

  • Unity
  • Blender
  • Arduino
  • C#
  • Python
  • In conclusion...

    This was a new and enriching experience for me. This was my first foray into the field of Virtual Reality and Real-time systems. I would like to thank Dr. Mayank Mehta for giving me the opportunity to work on this project. Mr. Chinmay Purandare, for for being an able and helpful mentor.


    Read the Report