Can someone direct me to some information about how to do the following:
I have a database of about 10,000 trading cards. Given an image from a iPhone/Android/WebCam, I'd like to find the closest matching trading card. I looked up some basic Haar based image recognition, but that requires training with many images with generally the same content. My program would have 1 perfect image for each possible match.
Now the input won't be scans, so the routines would need to handle scaling/rotation and cropping of some sort. But the entire image should fairly visible.
Any ideas, or thoughts? This is my first foray into the world of OpenCV, and it's interesting to say the least.