Are there reputable citations that the 3d facial reconstructions actually work? I know I’ve seen cases where a person has been identified after a reconstruction, but does the success rate fall better than a near random guess?
This might be hard to answer. We may be able to see how many successes we’ve had…but the failures can’t really be assigned to inaccurate forensic reconstruction.
I guess, I’d like to see evidence where an artist worked with several skulls where the individual was known (but not to them), to see how ‘good’ they were are representing reality.
Or maybe: several artists each create a model from the same remains with the assumption that they would all look pretty close to each other (and of course to the real person).
Have any studies like this been done? Do we know how good these artists are at consistently representing the true image of what a person looks like?
National Geographic commissioned two reconstructions of King Tut’s face, based on CT scans. One of the teams knew who the subject was; the other team didn’t. The two reconstructions look pretty similar.
How can they tell how thick to make the flesh? A person’s face can change appearance from a loss or gain of weight, but I assume that doesn’t affect the skull.
Actually, this is a very interesting question. I recently watched a couple of cop shows, and if these are to be believed, the result of the reconstructions is always good enough that the deceased’s blind grandmother could recognize them.
The links posted by “Northern Piper” above are in my opinion not very good, in that to me they look vastly different due to camera angles and degree of finish. I don’t think they are comparable.
In my opinion what would be required is that the same skull be given to about a dozen reconstruction experts, and their results compared side by side; even better if they have a good picture of the dead person to compare to as well.
Has this ever been done?
As a minor diversion: I am reminded of a similar test done some years ago of criminal profilers; if I remember correctly the profilers failed the test dismally, but when compared to average street cops, the cops beat them hands down. (no cite)
Muscle, skin and connective tissue are fairly standard in depth. Dissection of cadavers yields measurements according to origin of the body, and averages are made. So if they find a skull in northern Sweden, they’ll make use of average modern nordic measurements.
Fat poses a problem, but it highly unlikely that an ancient person would be fat in the sense that we know it. In any case, if they have most of the body, they can make an estimate of weight and make adjustments for fat based on that.
There is science behind forensic reconstruction of faces, but there is unquestionably speculation and guesswork as well, because there are certain things about the face that just can’t be told by bone structure. It’s really hard to quantify something like facial reconstructions, since there’s no real standard objective measure of how much something looks like somebody. Since the goal is only to create a resemblence that a friend or family member recognizes after seeing it in the news, “success” isn’t so much recreating an exact likeness of the deceased but rather in making enough of a a resemblence to attract the attention of people looking for someone they know to be missing. By that measure, we know that facial reconstructions have been successful, as people have seen them and been reminded enough of loved ones to contact authorites. If you want to see how much a recreation can look like the actual person, google Dorothy Ray Howard and Goldie Thornsberry. In the latter case, the reconstruction is really, really rough compared to the photo of the actual person, but it was close enough that a friend saw it on the news and contacted the police, leading to a positive ID.
The actual magazine article had several photos of each, from different angles; all told, they looked pretty similar, given the different mediums used for the two reconstructions.
In the last 40 years, or so, I’ve seen in the newspapers about 10 cases solved by using this method. They always show the reconstructed face and a picture of the victim. The reconstructed faces always looked, to me, like the same one as the previous case, and not like the victim. Always something that looks like an idol/statue or something from a Pre-Columbian archeological dig.
Everybody on the media swears that it is like looking at the person in real life. Pretty sad. I think it’s a bunch of malarkey.
Well, perhaps I should have said " ‘solved by this method’ ". I think that the real bodiesof the alleged identified are still in their shallow graves, but, the prosecutors, et al, have chosen to say ‘Miracle of Miracles! Through the wonders of Modern Science, the bodies have been identified!’
And, of course, ID’ing a body doesn’t really much *solve *a case. Unless the stiff is in the trunk of my car. And, then, the ID factor isn’t necessary, etc…
Whole point being, the forensic scientists found a skull, did their hocus-pocus, looked at some pictures, and used creative visualization to project the image onto one of the missing persons, and, voila,CIS!
Forensic reconstructions are rarely validated in studies because they can’t possibly replicate how they are used in the real world, where someone that KNEW that person recognized them.
I know these can work because I’m a forensic artist ----if you couldn’t tell by my title! and have seen them work time and time again.
The thing is, the reconstruction needs to be in a database 24/7 for searching, and the family needs to be looking. Also, just a bit of reality here, if every missing persons and fugitive’s DNA were in the National MIssing Person DNA Database, then we would have a fraction of the UIDs that we do now. I would happily be put out of ta job it that were the case. We do facial reconstructions (“facial approximations” is actually the more correct term) because they are the last chance that person has to be ID’d. after DNA, dentals and fingerprints have come up empty.
Here’s a very recent ID of a woman, I know the artist that did this, and she was ID’d after the family saw his approximation. This works.
and have seen them work time and time again.