I am asking myself either it would be possible to train a neural network to create one image out of a video.
My question is not to select one image inside the video,
