Escher and the Droste effect – WebGL fragment shader
1–10 of 13 posts
Re: Escher and the Droste effect – WebGL fragment shader
#2I don't have anything intelligent to say about this - but this for me is one of those moments where I feel positively in awe of how next level some people are.
Re: Escher and the Droste effect – WebGL fragment shader
#3Excellent! Thanks for sharing!
Re: Escher and the Droste effect – WebGL fragment shader
#4This is really cool, since Portal I've been fascinated by the concept of creating optical illusions and optical phenomena with programming. While I believe that my mathematical knowledge is capable of comprehending this, I don't have any experience with the languages needed to program them. I'd like to though.
Re: Escher and the Droste effect – WebGL fragment shader
#5I thought that this gif https://lh3.googleusercontent.com/-Izdubw7RJWk/URK2tZdqxzI/A... was already impressive, but this is just completely incredible.
Re: Escher and the Droste effect – WebGL fragment shader
#6Here is a mathematical reconstruction of the print with several fly-throughs: http://escherdroste.math.leidenuniv.nl/index.php?menu=intro
Re: Escher and the Droste effect – WebGL fragment shader
#7I've seen this painting before but didn't think much of it. Now that I can see what's going on, I am dumbstruck with how abstract and novel this concept is.
This spacial reasoning ability is so amazing.
Re: Escher and the Droste effect – WebGL fragment shader
#8Why is there visible angular distortion when you zoom in? It shouldn't happen if you do the Droste effect correctly, like this: https://www.youtube.com/watch?v=9WHdyG9mJaI#t=0m43s
Re: Escher and the Droste effect – WebGL fragment shader
#9I thought that this gif https://lh3.googleusercontent.com/-Izdubw7RJWk/URK2tZdqxzI/A... was already impressive, but this is just completely incredible.
The gif is still very impressive. If I recall correctly the paper said that they did this by finishing the white hole in the original piece so they could do the conformal mapping completly.
More impressive is that Escher did this without a formal math training
Re: Escher and the Droste effect – WebGL fragment shader
#10Why is there visible angular distortion when you zoom in? It shouldn't happen if you do the Droste effect correctly, like this: https://www.youtube.com/watch?v=9WHdyG9mJaI#t=0m43s
My bet is that he is using a continuos transformation and the guys who first did this (at least as far as I know) reproduced and completed the original transformation grid that Escher did.
http://escherdroste.math.leidenuniv.nl/index.php?menu=symmet...
The transformation is a conformal map so it should not deform angles. Other explanation is that the fragment shader is losing information in the process and that float numbers are not precise enough.
Still this is very impressive