Processing the video as separate images could take quite a long time as there are many hundreds if not thousands of frames to be processed. I’m not sure this is the best way to go?
The sample clip is taken from the “extended” version of Blotto. But as far as I’m aware, there are no clean or clearer extended video versions available apart from the video that I have. There are clearer videos of Blotto, but they are shortened versions only as far as I know, and don’t include extended scenes.