Live data from Hacker News

VideoGigaGAN: Towards detail-rich video super-resolution

videogigagan.github.io

81–90 of 243 posts

Re: VideoGigaGAN: Towards detail-rich video super-resolution

#81
post #74

This is great for entertainment (and hopefully the main application), but we need clear marking of such type of videos before hallucinated details are used as "proofs" of any kind by people not knowing how this works. Software video/photography on smartphones is already using proprietary algorithms that "infer" non-existent or fake details, and this would be at an even bigger scale.

Funny to think of all those scenes in TV and movies when someone would magically "enhance" a low-resolution image to be crystal clear. At the time, nerds scoffed, but now we know they were simply using an AI to super-scale it. In retrospect, how many fictional villains were condemned on the basis of hallucinated evidence? :-D

Re: VideoGigaGAN: Towards detail-rich video super-resolution

#84
post #46

This is amazing and all but at what point do we reach the point of there is no more “real” data to infer from low resolution? In other words there are all sorts of information theory research on the amount of unique entropy on a given medium and even with compression there is a limit. How does that limit relate to work like this? Is there a point at which it can say we know it’s inventing things beyond x scaling cons…

> This is amazing and all but at what point do we reach the point of there is no more “real” data to infer from low resolution?

The start point. Upscaling is by definition creating information where there wasn't any to begin with.

Nearest neighbor filtering is technically inventing information, it's just the dumbest possible approach. Bilinear filtering is slightly smarter. This approach tries to be smarter still by applying generative AI.

Re: VideoGigaGAN: Towards detail-rich video super-resolution

#86
post #74

This is great for entertainment (and hopefully the main application), but we need clear marking of such type of videos before hallucinated details are used as "proofs" of any kind by people not knowing how this works. Software video/photography on smartphones is already using proprietary algorithms that "infer" non-existent or fake details, and this would be at an even bigger scale.

Like Ryan Gosling appearing in a building https://petapixel.com/2020/08/17/gigapixel-ai-accidentally-a...

Re: VideoGigaGAN: Towards detail-rich video super-resolution

#87
post #58

Video quality seems really good, but limitations are quite restrictive "Our model encounters challenges when processing extremely long videos (e.g. 200 frames or more)". I'd say most videos in practice are longer than 200 frames, so lot more research is still needed.

At 24fps that's not even 10 seconds. Calling it extremely long is kinda defensive.

The average shot length in a modern movie is around 2.5 seconds (down from 12 seconds in 1930's).

For animations it's around 15 seconds.

Re: VideoGigaGAN: Towards detail-rich video super-resolution

#88
post #59

This is great. I look forward to when cell phones run this at 60fps. It will hallucinate wrong, but pixel perfect moons and license plate numbers.

Just get a plate with 'AAAAA4' and blame everything on 'AAAAAA'

Even better, get NU11 and have it go to this poor guy: https://www.wired.com/story/null-license-plate-landed-one-ha...
Post reply on HN