Inspiration
In the 1990s, researchers introduced a famous paper, 'Tour into the Picture.' Their algorithms allowed people to use perspective and vanishing points to convert a single 2D image into a 3D video-game world. With the explosion of VR in recent years, it felt natural to create a tool that can produce VR arenas or even large generated worlds from just a few photographs.
What it does
Users can upload an image at vr.mbrodie.com. This sends a request to a server that processes the image and produces a downloadable unity codebase that will run on Google cardboard, Oculus etc.
How I built it
I scraped out a simple doom-like game, and introduced my own functions to design the virtual world. I used javascript/html/css to create the landing page where users can upload and convert an image into a 3D VR experience. I used python and OpenCV for running vision algorithms and compiling the code.
Challenges I ran into
I hoped to learn about and demo the project on an Oculus Rift. Unfortunately, Oculus does not yet support Mac drivers. I also had never used Unity, which was challenging and fun use.
Accomplishments that I'm proud of
I have a landing page, an algorithm, and a process that holds promise for future enhancements in VR.
What I learned
I learned Unity and improved my ability to hack together projects using a variety of tools and technologies.
What's next for VR into the Picture
Stitching together MULTIPLE pictures, or allowing users to build a world with portals connecting image to image.
Log in or sign up for Devpost to join the conversation.