This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
The interest of ROI is to speed-up computation, therefore, a triangular ROI seems problematic. But, if you are only interested to have a part of your image, you could create a ROI bounding to your triangular area, and add a mask, of the ROI size, that specify the pixels to keep. After, just copy the ROIed image with the mask as third parameter, you will get the triangular ROI-like. But again, the interest has to be specific (hide parts of the image, prevent from annoying moving background,...) but not speed-up has you involve an extra copy of data.
