Augmented Reality is becoming a bright spot in the horizon as exciting technologies like smart glasses are being developed. Our team was really interested in that field and wanted to develop technology in that area. Our goal is to use hand gestures to make daily tasks more exciting, convenient and to-go without ever having to carry anything with you. We have a vision where in the future, two people can walk up to each other and electronically write a check by merely writing the information in the air. We see a future where receipts and bills can be electronically signed by just giving your signature in the air. There's a future where artistic creativity can bloom as artists utilize AR technology to fully express themselves. This was the motivation behind developing AIR.

Working on this project taught us a lot about computer vision. We heavily utilized OpenCV and learned about various image and video processing techniques such as background subtraction and color thresholding. We also learned to work with Google's Vision API to develop a robust system for handwriting recognition. One big challenge we faced was stabilizing the written output as noise in the video capture would result in very jagged lines being drawn across the screen. We solved this by creating moving averages and detecting if the new computed finger location was too far away from the previous average location.

Our team is very proud of what we managed to develop during this two day hackathon, and we're all very glad we got the opportunity to be able to contribute to the open source community.

Our team consisted of Freshman ECE Majors from the University of Illinois at Urbana-Champaign

  • Joseph Rejive -Manav Modi -Chirag Rastogi

Built With

  • checkbook.io
  • google-cloud-vision
Share this project:

Updates