Decode h.264 video and convert to cv::Mat

Hi, I’m using OpenCV 4 (Compiled with MSVC 15-64bit + CUDA 10) and VideoCapture class to decode the h.264 stream, but there is about 5 seconds delay in displaying frames. unfortunately my camera doesn’t support mjpg and I have to use h.264. I’m wondering if there is a way to use GPU for decoding the stream?