As I know, Homography (projective transformation) in computer vision can be used to detect object in images but all the object I've seen are plane objects. Does Homography only work on a planar surface surface object? Or It can detect any kind of objects? I'm asking because I tried to detect non planar surface image and it didn't work.
I think you have a conceptional error here. A homography does not detect anything, it just describes a mapping from one plane to another. In the case of the book in your example, the homography tells you how to map the plane of the front cover of the book to the image plane. If you have a non planar object, then there is no such homography.
Homography does not detect any object. The link you posted shows object detection through feature matching, which can also be used to compute homography. If you want to do object detection you should look into feature matching, HOG, or the newer and more effective deep-learning systems such as convolutional neural networks.
Asked: 2015-06-04 07:00:46 -0600
Seen: 991 times
Last updated: Jun 05 '15
if you got 3d model points, and 2d correspondances, it might be a case for solvePnP
What do you mean by 3D model points? can u please clarify?