I might have worded this a bit too complicated so on request i will try to explain it a bit more simple and add some better explanation.
In order to understand how stereo footage such as this is shot usually 2 satellites are used, each carrying a camera, The reason for this is to increase the distance between the cameras so we can get a 3d effect. Same as our own 2 eyes work but we usually look at objects way closer and once we look at something that is very very far away the 3d effect is to subtle to notice, hence would beat the purpose to have 2 cameras that are too close to each other on a satellite that captures footage of distant object for stereo view.. It might of course be that there are satellites that have 2 cameras but it is all the same because you do need 2 cameras. a digital camera has a sensor, the photosites of the sensor capture the photons and measure the values, i wont go into detail how it works as this would be a very long text but long story short: the sensor creates a noise pattern due to the fact that each photosite is constantly capturing photons,the noise pattern is absolutely unique and completely different in each frame, even if the camera and object are not moving at all. the only noise patterns that are persistent us called pattern noise , it usually occurs when a sensor gets pushed to the upper ISO limit, this type of pattern noise usually looks like long lines on the screen, it does not affect the whole screen and does look nothing like this.i work with highend cinema cameras both with CMOS and RGB sensors. it is not possible for 2 different cameras to create a matching noise pattern, it does not matter if they look at the same scenery, nor it does not matter if the cameras are from the same manufacturing line. it is simply technically not possible for the sensors to be hit by the exact same number of photos, hence noise changes in every frame.even if you would shoot super highspeed footage with one cameras, in each sequential frame the noise pattern would be completely unique. if you overlway one side of the 3d video with the other side you will see that the pixels of the pattern do not match, the pattern looks similar but not identical. this is because the stereo view was generated after the footage was recorded, in order to generate a stereo view the video must be distorted on one side, otherwise you will not get any 3d effect and because the video was distorted the pixels no longer match.You can however clearly see that the random pattern on both sides looks very very similar.this is absolutely not possible in real stereo footage that was shot on 2 different cameras.it is technically absolutely not possible and since this happens in every frame you can absolutely rule out coincidence.
Thank you for reading.
I captured the video originally posted on youtube in 2014 and had a closer look at it.i applied strong sharpening to make the noise and compression artifacts become a lot more visible.i did some overlays to compare the sides and i quickly noticed that the mix of noise pattern and compression artifacts looks pretty much the same for most of the footage (i say most because i did not go over the whole video frame by frame)https://web.archive.org/web/20140827052109/https://www.youtube.com/watch?v=5Ok1A1fSzxYhere is the link to the original video
if you wonder why the noise pattern is not an exact pixel match it is easy to explain. since you can see that the image is stereo it simply means that the 3d effect was generated in post, hence areas of the image have shifted to create the effect. also rescaling and repositioning and ultimately re-encoding the video will add distortion but you can still see the pattern very clearly. There are multiple ways to create a stereo image and this particular video has no strong 3d effect . This can be achieved by mapping the image/video to a simple generated 3d plane with extruded hight for the clouds. There are also some plugins that will create a stereo effect for you.
i have marked 2 areas for you, you can see the very similar shapes there. these are of course not the only 2 areas, its the whole image in all the frames but it is easier to notice when you start looking for some patterns that stand out. the patterns are of course in the same area on both images. you can spot a lot more similar patterns just by looking at the image.
– only look for the noise and compression artifacts, those change with every frame and not part of the scenery.
What does it mean? It means that this video was doctored and that someone did put some effort into making it appear more legit. that is all. There is absolutely NO WAY that 2 different cameras would create the same noise pattern and the encoder would create the same artifacts. even highspeed images shot on a completely still camera will not produce the same noise patterns in sequential frames.
feel free to capture or download the originally posted video and do your own checks.