How fake ufo videos are made

The original video got 324,000 hits while the reveal has only 111,000.

This is pointless statistic... aimed at influencing peoples minds.

Of course the first video has more views because it goes viral with excitement of this amazing footage, people will hear the buzz about it then watch it.
the second video is just a reveal, once people know it's fake they don't need to see how it's done.

Personally I think it's a high number of views, if a third of people could be bothered to watch it.

originally posted by: Pinke
a reply to: charlyv

This technology is already in place with some DSLRs to trip a flag when an image is edited. Unfortunately it won't change much as most people compress an image or video before uploading it anywhere. Being able to indicate an image is modified throughout a certain workflow isn't the issue often ... it's knowing the workflow, knowing how to confirm the results of that workflow, and being able to express that to the general public.

My time on ATS has me convinced that the general public will never reach a level of interest or education to accept report results without coming up with elaborate reasoning as to why an image could 'still be legit.'

I was once actually accused of inserting a helicopter into a UFO video to debunk it.

Not what I am talking about. This new method checksums the original image and encrypts it into the video fields. The keys are public and private and based on a chip in the camera, as well as a users passkey. The idea is that the only proof can come from an original digital image taken by that camera.

a reply to: charlyv

We're talking about the same thing, I'm just avoiding jargon. For the discussion at hand we don't need to refer to the exact method of authentication (there is more than one proposed method in this instance).

The issue is that on the images we're talking about that we look at on ATS the CRC won't check out anyway. Whether we're talking digital watermarks compared using CRC or other methods involving naturalized image statistics compared to the camera sensor (which have been available for decades) neither will actually resolve any of the debates we have on ATS.

In signal processing we have been in the position that for years we can use the original camera and lens to authenticate an image within a high degree of likelyhood. In fact, we can often discount an image's authentication immediately by comparing the story we receive to the resulting image. We're still left with the same issues on ATS.

1. Most people don't comprehend natural image statistics or CRC and therefore don't have a clue what you're talking about
2. Evidence of alteration is not evidence of fakery regardless of how silly the image looks, so even if they do 'get it' they can explain it away ... 'oh you know, I just compressed it another colorspace, woopsie'
3. The processes you're talking about have already been circumvented; regardless of this being time consuming and rare people can invoke it when they want to
4. The process you're talking about generally won't discern between innocent edits and malicious ones

Don't get me wrong, I actually have a degree of faith in these methods. They work in most situations, but the above points still stand.

Geez,whatever happened to the old toss the hubcap in the air ploy?