VideoTomb obfuscates / encrypts videos so humans won't be able to see the content, and they'll probably pass video restriction detections as well.
For an example output video, see Sample.mp4 in the Sample directory. The buck bunny video is used as input. As you'll see the mask image is the only thing distinguishable from the output Sample video, while underlying is the buck bunny video.
The videos can be watched with the HTML/JS decoder in VideoTomb/player directory, which will embed the video player of the upload site in an iframe (can be Youtube), with the encoding mask on top.
Mono support is untested, currently working on it, but depends on Accord.
Noise hash functionality as an alternative to image mask is still in the source, but the Javascript side has been lost partially. I've decided to exclude it, but maybe I'll get back to it. This would allow to supply an encryption/decryption hash in the url. The original Javascript code has been lost and only the Uglified version remained. I've put it through a beautifier but I can't seem to find why noise hash functionality doesn't work.. Sorry!
I tried to clean up the directories and files a bit, but am no longer actively working on the project.
Perhaps I'll pick up work later on.
There is an option to encode sound with noise cancellation. The way this works is that left and right channels have random noise added and only output the video sound when joined together when enabling mono sound on Windows or Android. Works wonderfully but would be easily detected by video restriction detection.
The encoder has trouble with longer videos which I have been unable to solve.
Licensed under MIT.