Do image and video quality metrics model low-level human vision?

Re-assign community