Live data from Hacker News

LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

studios.disneyresearch.com

21–30 of 36 posts

Re: LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

#22
post #16

This reminds me a bit of a project I am currently doing where I swap adjacent pixels in the last image of short video sequences until it resembles the first image: https://www.instagram.com/marekgibney/reel/DILksFYNSkE/

That's quite an extraordinary effect. There's a point where enough pixels have been swapped that you can see the original, and it jumps out, and thereafter the image is quite surreal before eventually being 'tinted'.

(Also shows the video compression in the final frame well, because it doesn't match the first frame. Just found that interesting.)

Re: LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

#24
post #16

This reminds me a bit of a project I am currently doing where I swap adjacent pixels in the last image of short video sequences until it resembles the first image: https://www.instagram.com/marekgibney/reel/DILksFYNSkE/

That's quite an extraordinary effect. There's a point where enough pixels have been swapped that you can see the original, and it jumps out, and thereafter the image is quite surreal before eventually being 'tinted'. (Also shows the video compression in the final frame well, because it doesn't match the first frame. Just found that interesting.)

Thanks. Yes compression arteficts always become very visible. Especially in low-detail areas like the sky.

Re: LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

#25
post #21
post #15

It looks like this could be used for some form of steganography. You need the right mirror form for decoding.

laplacian pyramid warping or just anamorphosis? 'anamorphic encryption' schemes have been used for hundreds of years.

Can you give a specific example? The term “anamorphic encryption” was used in a paper from EUROCRYPT 2022 and I don’t see any prior formal study of the subject. A follow-up paper from PoPETS 2023 establishes that most of the widely-used cryptosystems introduced over the past few decades are indeed capable of anamorphism but nobody used the term “anamorphic encryption” until the last few years.

To say that they have been used for “hundreds” of years seems like a sweeping generalization in need of better context.

Re: LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

#27
post #15

It looks like this could be used for some form of steganography. You need the right mirror form for decoding.

When I saw the video I wondered how many layers of refraction you could achieve while maintaining coherent images. The pixelated style they should might be the easiest use to develop this from.

Re: LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

#28
post #24

Earlier quoted context omitted.

That's quite an extraordinary effect. There's a point where enough pixels have been swapped that you can see the original, and it jumps out, and thereafter the image is quite surreal before eventually being 'tinted'. (Also shows the video compression in the final frame well, because it doesn't match the first frame. Just found that interesting.)

Thanks. Yes compression arteficts always become very visible. Especially in low-detail areas like the sky.

What is the algorithm you are using for replacement / swap?

Re: LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping

#29
post #24

Earlier quoted context omitted.

Thanks. Yes compression arteficts always become very visible. Especially in low-detail areas like the sky.

What is the algorithm you are using for replacement / swap?

I have written up a description here:

https://www.instagram.com/marekgibney/reel/DDezhSisD4J/

Post reply on HN