Augmented Reality: Marker Recognition Only

Question

Think about the usual implementations of augmented reality: there is one or more markers, the software recognizes markers and usually shows something on top of them.

I'm only interested in the first part. I need an algorithm, component, structure or class to determine if a given marker is present or not in a photo.

Where to begin? I prefer the component for Java or C ++, but I am not limited to these languages.

Answer 1

I think you should see opencv detectors: Function detection (there is an example, find_obj.cpp sample comes with an opencv package). The most popular algorithm is described and implemented in ARToolKit (just google it).

Answer 2

I'm not sure if this is useful or not, but for Java you can take a look at ZXing . This is a barcode reader library, but the task of reading QR codes is to recognize 3-4 markers in a barcode. You can find his approach repeatedly.

Answer 3

OpenCV ++ ( Python) .

Answer 4

. OpenCV, .