maushammer
Three webcam test
This is a test of the actual webcams I want to use. I 3D-printed a holder and connected them up to a netbook. The netbook won't provide enough processing power for the real thing, but it is good enough to record so that I can evaluate the performance on a real computer.
I'm evaluating the cameras based on exposure time (my first try managed 1 fps indoors), field of view, focus (sharp corners are necessary to detect features) and jello. Yes, jello, otherwise known as how the picture becomes distorted due to motion and vibration.
I worry about this year's course - It will be at the reservoir and I'll have to align the cameras so they see either the background (kinda sparse - no big building to look at) or the cones (all look identical - could be confusing) and not the people (I expect many more people to be visible - they are not reliable waypoints!!)
Three webcam test
This is a test of the actual webcams I want to use. I 3D-printed a holder and connected them up to a netbook. The netbook won't provide enough processing power for the real thing, but it is good enough to record so that I can evaluate the performance on a real computer.
I'm evaluating the cameras based on exposure time (my first try managed 1 fps indoors), field of view, focus (sharp corners are necessary to detect features) and jello. Yes, jello, otherwise known as how the picture becomes distorted due to motion and vibration.
I worry about this year's course - It will be at the reservoir and I'll have to align the cameras so they see either the background (kinda sparse - no big building to look at) or the cones (all look identical - could be confusing) and not the people (I expect many more people to be visible - they are not reliable waypoints!!)