This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
Short answer: it's a pure virtual function, so it has no utility and the idea is to provide a common function interface for the inherited classes.
Longer answer: ShapeTransformer is a virtual class with no functionality. It's a common base for the Affine and ThinSpline transformers, so OpenCV's algorithms can use both classes using the same code (hence the "pure virtual" function).
That said, you can avoid to use directly either the AffineTransformer or ThinPlateSplineShapeTransformer classes, it's more simple to use the geometric image transformation functions, like WarpAffine, WarpPerspective, Remap, LogToPolar, Undisort, etc.
