Computer Vision - Feature Detection and Tracking - Tracking a RC car in a video

I would like to be able to track three different features of a RC car in an avi video. The camera is fixed. I hope to be able to calculate the actual position of the car once I am able to track the car in the video.

The main problem is with not knowing much about feature detection and tracking. I have the student version of matlab but don't have the computer vision toolbox. Is there a way I can do this feature detection and tracking without using the computer vision toolbox?

Answer 1

[Geoff]

Of course, but you might have to implement a bunch of complex algorithms yourself. If you're not familiar with vision processing, you might struggle.

What features on the car are you talking about? Do they have a unique shape and/or colour? Are they always visible? As for recovering the position of your car, your success will depend on how well you are able to calibrate your camera and/or how flat the ground is, how high your camera is, lighting conditions etc....

Simplest way to track a moving object in a stationary camera is by frame differencing. But the complexity of your solution depends on what else is moving in the scene. I would start with blob tracking, and if that doesn't suit, maybe optical flow.

It might help if you post up some frames from your video.