If I encode a video in Adobe Media Encoder or Adobe After Effects, is there some way, either in one of the programs or a plug-in or cheap/free utility to see some sort of "signal to noise ratio" value (a single value) for the encoded video compared to the original?
I'm not totally sure it's the right term but I want something that tells me how close a particular encoded video is to the original source video. eg. it would let me know if, because of artefacts/noise introduced eg. because you gave it a low bitrate or other parameters that one encode was very unlike the original source video, and that it would tell you that another video was identical or was almost identical to the source (eg. no compression artefacts or other differences to the original introduced).
I'm looking for something that would give me a single numerical value for a video compared to another video - ie. encoded video compared to the original (signal-to-noise ratio?).
Thanks. But I'd much prefer it if there was a numerical value as the result (1 value for entire video), so I could compare different encodings to the original (source) video and the numerical value for each would let me know which were the closest matches.
video A encode is 100% the same as source video
video B encode is 95% the same as source video
video A. signal to noise ratio (compared to source video) is 99999:1 [video A is very close to source video]
video B. signal to noise ratio (compared to source video) is 1:1 [video B is very unlike source video]
Maybe I've not got the term quite right but I'm looking for a a simple way of comparing different video encodings of a source video, to determine what particular video formats/encoding formats/parameters produce the most similar video to the original source videos (without doing a visual check for encoding artefacts).
The difference matte would give a difference for each frame as image frames (a video) rather than 1 numerical value for the difference between the entire videos. The one numerical result per encoded video would make it a lot easier to compare lots of different video encodes, with different parameters (against different source videos) - without visually looking at each frame produced by the difference-matte method.
Another thing to factor in is that the mathematical signal to noise ratio doesn't account for the psycho-visual compression that is acceptable to the human eye. You potentially can have a lower SNR but the image is visually worse than another of a higher SNR.
It is better than nothing though.