Now I am testing cvtcolor() function of Python API of CVCUDA lib. However, I do not know how to select an exact code for converting from RGB to BGR image. The below is my test code:
Sorry I won’t be able to work with screenshots. I need to copy text out of your screenshot and I can’t do that. I’m not going to try typing it. Please make it easier for others to help you by posting text as text on this forum, not as screenshots.
In a nutshell, I would say that the first argument in your call is a nv image codec object, and cvcuda is expecting something else there. You created nvtensor right before this line of code. Why are you passing nvimg instead of nvtensor ?
Try this:
cc = cvcuda.cvtcolor(nvtensor, cvcuda.ColorConversion.RGB2BGR)