I have a video from a non-moving camera. It has recorded a sign. For the most part this sign does not move on screen. However, by the oddities of the camera recording, each frame has captured the wording in a slightly different "blurriness" - no one frame captures the sign consistently. Could individual frames be overlaid on each other in such a way that a more advanced technique could be used to infer what the letters on the sign are? I've already tried just laying the frames on each other and adjusting opacity of all of them to get a kind of "average". That didn't work.