Talk:FingerSpark

Proposal Review

Your idea is sound and feasible, but a bit loose on the details. I have a few questions:

1. What is the backup plan if the Raspberry Pi camera does not have enough resolution? In your project there is a direct relation between number of pixels, refresh rate, and how far you can have your hand from the camera. You can try using your laptop's webcam to guesstimate the limits and constraints in your project, since it will be weeks before you start acquiring images using the Raspberry Pi, and you cannot afford to wait that long to improve your design if the resolution is inadequate.
2. Once you capture the positions of the fingers in the camera's coordinate frame (i.e., once you have the X-Y coordinates of your pixels), what are you going to do with the information? Are you going to transform it back to the user's coordinate frame? In other words, what is going to be your demo at the end of the semester? Is it going to be a screen printing numbers with the coordinates, or is it going to be an actual application? If the answer is the latter then you need to add the application to your timeline, since it will take a significant amount of time.
3. Transforming from finger coordinates to gestures is a challenging task. Do you have an idea of the type of algorithm you are going to implement to solve that problem?

Also, your budget does not specify what needs to be purchased and what is already available. Please meet with your TA to discuss these issues and update your proposal as soon as possible.

Humberto (talk) 20:01, 12 February 2016 (CST)