"Arturo Castro's real time face substitution technique"

Link/Videos: http://www.notcot.org/post/43233/

An HN comment I found very interesting:

Real-time virtual puppeting has been done in movies/television and research for a while now and yes, it can easily fool people.

A professor friend of mine Jeremy Bailenson at Stanford actually uses the Kinect to track facial movements and uses 3D models of others to create puppets in real-time. Even more interesting, he can morph your face with the person you're video conferencing with to create a feeling of commonality in them.

He actually wrote a book on it called Infinite Reality [1] which talks about all kinds of ways people will probably get manipulated in the future. He talks about things like mirroring movements (which he can do automatically in a video conference), looking into the eyes of every participant in a group video conference and other really interesting psychological hacks.

[1] http://www.amazon.com/Infinite-Reality-Avatars-Eternal-Revolution/dp/0061809500/

From: http://news.ycombinator.com/item?id=3022929

Hacker News discussion thread: http://news.ycombinator.com/item?id=3022479

New Comment
5 comments, sorted by Click to highlight new comments since: Today at 12:13 AM

So much for technology helping us be more verifiably honest: http://hanson.gmu.edu/moretrue.pdf

Now it looks much more like an arms race.

I see it as a cat-and-mouse game, like the spam problem.

Or even better is the problem of Photoshopping things which people have come up with some I guess fairly good tool to counter:

“Error level analysis (ELA) works by intentionally resaving the image at a known error rate, such as 95%, and then computing the difference between the images. If there is virtually no change, then the cell has reached its local minima for error at that quality level. However, if there is a large amount of change, then the pixels are not at their local minima and are effectively original.”

From http://errorlevelanalysis.com/

Also people could be more skeptical of video from an unfamiliar source, as they must be getting with photos (magazines covers, especially US beauty-oriented ones) and movies (special effects getting better and better, the young Jeff Bridges in Tron 2.0 didn't fool me but I remember reading a comment or two about people saying they were fooled).

My link to the ELA page and inspiration about using one's judgment came from these not-so-technical articles that cam up in a quick Google search:

http://lifehacker.com/5644259/how-to-detect-a-photoshopped-image

http://www.ratchetup.com/eyes/2007/04/detecting_photo.html

I've heard ELA isn't actually that good. For example, notice that they never actually verified what was photoshopped in the example picture.

Also, based on the description of how it works, I'd guess that it only works if the image is compressed before it's photoshopped. This would seem unlikely if the creator has access to the original image.

I'm not remotely an expert in this. The first paragraph is just me echoing something I've heard from someone else, and the second could easily be me misunderstanding how it works.

...looking into the eyes of every participant in a group video conference...

Unless you have multiple cameras, looking into only one of their eyes would be the difficult part. You'd normally be looking at the screen instead of the camera, so making it look like you look into the camera would be helpful, but that has nothing to do with a group conference.

This is going to revolutionize the way people have sex! On the Internet!