We are trying to solve an interesting problem. We have video streams (screen casts) of our writings stored in digital format. Some thing like this:
Video_id x_coordinate y_coordinate time_stamp color_code
In total we have some 50 videos from 5 people totaling up to 200 minutes.
We are now trying to build an application in R, which can train itself on these videos and then can classify which instructor is writing during the video for future videos. Here are some of the questions we have:
- Is R a good choice for this problem? Or should we use Python?
- If it is, what could be a good strategy to build this classifier?
- Which algorithms and libraries should we use? Is this something where Neural networks can be applied?
Finally, if any of you are aware of any tutorials which can help us build this application, it would be of immense help.