Developer Spotlight: Pranava Singhal

What was your first experience with Spectacles?

I first got involved with Spectacles at Immerse the Bay, Stanford's XR Hackathon. Our team was inspired to create Immersive Ink for the Spectacles because we wanted to unlock the potential of augmented reality for making reading more immersive, engaging and accessible.

What was the impetus behind the creation of your Immersive Ink Lens?

We wanted our users to feel like the characters in their books were jumping out of the pages to narrate their stories. Spectacles helped us achieve this with its powerful combination of optical character recognition, high quality renders that even work outdoors, and built-in AI tools like GPT and text-to speech.

Did you run into any roadblocks along the way?

Some challenges that we faced when building Immersive Ink were the spatial misalignment between the camera feed, and what the user sees, which made OCR capture a little tricky. We worked around this by showing users a small overlay with the real-time camera feed. Another challenge was the long waiting time for text-to-3D generation APIs. We circumvented this by rendering a preview mesh of the object as a 3D analog to the buffering animation. This was to keep the user engaged while textures were still being generated.

Any advice for students looking to create for Spectacles?

My tip for other students building with Spectacles would be to get creative with the extensive capabilities within Lens Studio, and to make use of the built-in GenAI Suite to speed up development.

What's next for you and your team?

I'm excited to continue building on Immersive Ink with my Hackathon team to make the experience more seamless by rendering objects in real time and supporting multiple viewers. I'm also looking forward to building something really cool at MIT Reality Hack.