Aug
03
The Helsinki Corridor
I have ventured outside Dallas by researching Lee Oswald in Helsinki. Don’t ask why, it was just something that I did for a laugh and started to become something quite interesting. See for yourself, my paper, 56 pages containing more than 130 links of which over 110 are unique and my foray in creating AI images based on the data and the photographs for reference.
Since this has almost nothing to do with Prayer Man I only post here about this chapter here once. The second edition of Prayer Man; More Than A Fuzzy Picture will be delayed since this paper is the beginning of a Cold War book I hope to have finished by end of this year. This paper is one of its chapters.
Hope you find it as interesting as I did.
View/download the paper at the DealeyPlazaUK website.
Apr
07
JFK Assassination – NBC Films – Prayer Man & AI – Part 3
This time I used Grok, at first I was like; mwah.
Wiegman in Grok Click pic to enlarge.
Wiegman in Grok, Click pic to enlarge.
Then added Darnell in the mix as well. This one looks better for the people in front of the steps than the ones on them, but overall this is a worthwhile attempt.
It may not help with Prayer Man at this point, but it does assist with other individuals.
Darnell sharpened in Grok. Click pic to enlarge.
If there are high res scans of these films then these would be even clearer.
Then you’d have an even bigger issue, and that is exactly why higher res images will not happen.
Apr
04
Time to mess about with AI, and the first attempts using Ideogram were fun! At first I thought I will just use a prompt of creating an image of Oswald being on top of the TSBD stairs.
The whole thing looks wrong but I couldn’t resist letting out a little laugh. The prompt is simply not good enough, and I will need to explore further.
Then I added one photo of the Wiegman film in for context and that made it all the more interesting if you check out the first four images below.
Following that I also submitted a Darnell still (images 5-8). I know everyone is all over but it kept its looks and character intact. Just interesting to try things around for a bit and see what can be created so easily already.
This could well become an issue with images being created and used in social media and believed as being genuine.
