We all come from very different backgrounds - Anton from physics, Bala in CS, and Brandon in math and economics. However, even with the huge differences, we all have enthusiasm for using technology to help people across the globe- regardless in what fashion. As such, we brainstormed for a few hours about potential projects taking into account the resources available to us here at LA Hacks and came up with KAR, which we think has the potential to legitimately help people in dangerous situations.
KAR uses the built in APIs within Kinect to identify and count the number of people within it's view range - from soldiers in the battlefield to firefighters in house fires.
We actually started off working with Microsoft's cognitive software to find people in the frame and their emotions (then send the information to Azure), but quickly realized that integrating all of those elements together would take much more time than one weekend. Additionally, So, in a way our greatest problem was the scope of the original idea, However, as we pursued development of our updated concept, we ran into tons of problems with library acquisition and declaration in the beginning sections of pretty much all of our code, but resolved them after hours of hacking!
In order to actually do something in the real world, we would need to refine the code that powers KAR and add in the augmented reality functionality (Hololens, Google Glass, etc). However, we kept this in mind while coding, so every step that we took was optimized for compatibility with later concepts, including facial/cognitive recognition to even further refine the concept, making use of Azure for a cloud-based platform for mobile applications of this technology (i.e. on a drone or R/C car) to proliferate, which ultimately would mean more lives saved.
Built With
- azure
- c#
- kinect
- microsoft-hololens
- microsoft-visual-studio
- native-kinect-apis
Log in or sign up for Devpost to join the conversation.