Hello,
I'm a little confused. Your method is ultimately about processing video, which should be converted frame by frame. But why can such methods produce relatively smooth video results without considering the temporal consistency of the video?
Or why not model temporal consistency.
Hello, I'm a little confused. Your method is ultimately about processing video, which should be converted frame by frame. But why can such methods produce relatively smooth video results without considering the temporal consistency of the video? Or why not model temporal consistency.