10 comments

[ 4.6 ms ] story [ 37.8 ms ] thread
I was really hoping this would be a paper describing how to "enhance" the image a la the CSI TV show[0]

[0]: https://www.youtube.com/watch?v=3uoM5kfZIQ0

I always wondered whether you could train a neural net to to do deep zoom.
I remember seeing this in the Wild Wild West[0] movie from 1999. It's curious seeing the history behind it.

[0] https://www.youtube.com/watch?v=6LFHZDXhUI8

Seriously though, doesn't it bother you that the primary gag is to repeat the same five words at each beat in the rhythm of that scene? It's such a lowest common denominator bit. This is sort of an odd crux to juxtapose, as a premise of art immitating life.

On the one hand, you have a plot device delivering some interesting sci-fi exposition to move the story forward, and on the other hand, you've got this squeamish every man wringing his hands from the vantage point of some supposed moral high ground, each ignoring the other, distracted by their own fixation.

Really makes you think.

Also used as a plot device in one of the amazing Lord Darcy stories, by Randall Garrett. A detective and his Watson in a medieval setting where magic is not particularly magical, but actually forensic science. They obtain a very clear image of the murderer from the victim using a spell, but it doesn't resemble any of the possible perpetrators.
I recall many years ago reading an article about how, with sufficiently powerful computers, it would be possible to reconstruct what a person in a photograph is looking at by enhancing the pinpoints of light reflected off their pupils.