Hi Nadex,
Could you clarify what you're trying to do? If I understand correctly, it sounds like you've already finished building the object detection system, and you'd like to add the functionality to track those objects between multiple frames of a video file. Is that the case?
I also have a few more clarifying questions, to make sure that we're on the same page:
-What kind of objects are you tracking?
-Why have you decided against pattern matching? Several threads asking similar questions found that pattern matching was their best bet, so knowing why it won't work in your situation will give us information about your needs.
-What do you mean by "screen vid[e]o capture"? If this is something you don't want to use, I want to make sure that we fully understand what you're referring to.
-What have you already tried? What results did you find when you tried those options?
In addition, screenshots of the object detection input and resulting images would also be helpful, so that we're aware of what size/shape/color of object and environment you're working in.
Of course, if you have other information that may be relevant, please let us know that as well, and we can go from there.
Thanks,
Kathryn K.
Technical Support Engineer
National Instruments
http://ni.com/support
