AI TECHNOLOGY RECOVERS 3D IMAGES OF THE SOMERTON MAN
With thanks to the Computer Vision Laboratory at the University of Nottingham, we have been able to create 3D views of the Somerton Man post-mortem and the plaster bust.
1. Post Mortem Photograph: 3D Video Images
The photograph in the video above is one that was reconstructed by SAPOL before its release on December 4th 1948. There has been much discussion about this image and following the extensive analysis done on this blog it is now generally accepted that the photograph was altered.
2. The Plaster Bust: 3D Video
Bear in mind that Jestyn like most of the population in Adelaide at the relevant time would have seen the published images of the man found on the beach at Somerton on December 1st, 1948. But, when she was shown the plaster bust she almost fainted to all accounts. It is believed that she instantly recognised the face of the man but would not acknowledge that fact. My view is that it is the plaster bust that shows a true representation of the Somerton Man and as you can see, it is quite different in a number of ways to the post-mortem image.
Followers of this blog will recall that in Clive's interview with Paul Lawson, he admitted that he used photographs of the man to create the plaster bust facial features. The question is, which photographs?
For those interested in the technology used, essentially it takes a single full-face image and using AI, it is able to generate a realistic 3D image of that face that can be manipulated.
The paper they produced, Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression, is available here:
Followers of this blog will recall that in Clive's interview with Paul Lawson, he admitted that he used photographs of the man to create the plaster bust facial features. The question is, which photographs?
For those interested in the technology used, essentially it takes a single full-face image and using AI, it is able to generate a realistic 3D image of that face that can be manipulated.
The paper they produced, Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression, is available here:
It's a great new tool to help in this case. The 'official' photo of the SM, as seen in the newspapers at that time, is nothing like the image from the bust. The 'official' photo of the SM seems to show different shapes for his eyes and his mouth is also shaped differently. The bust image seems to show different shaped eyes and mouth and, a younger(?) looking man? Why was there a new for a 're-constructured photo' in the newspapers, why not simply show the face as he was found? Clive
ReplyDeleteIt certainly looks like the eyes and the shape of the mouth are different from the 'official' police photo of the SM compared to the bust. The bust face seems to suggest a younger looking man? Clive
ReplyDeleteThe shape of the eyes and mouth are, seemingly different in the 'official' photo of the SM. Comparing the face of the bust, it seems it is the face of a younger looking man? Clive
ReplyDelete