Vesuvius Challenge • Hacker News Board

Re: Vesuvius Challenge

#11

Post by natfriedman » Wed, Mar 15, 2023, 6:00 PM UTC

SubiculumCode wrote: post #10

I do a lot of brain image segmentation in my research using multiatlas image segmentation, which involves diffeomorphic image registration from multiple labeled atlases...but the amount of curling in on itself of these layered sheets seems a daunting problem for a fully automated pipeline.

It seems quite possible that the solution isn't fully automated. N is in the hundreds. And modern AI does, in fact, involve quite a lot of hand crafted data...

Re: Vesuvius Challenge

#12

Post by localplume » Wed, Mar 15, 2023, 7:45 PM UTC

Reminds me of Kuzushiji recognition with ML, transcribing historical Japanese documents. Both are my favorite applications of ML: deciphering the past. This is really damn cool.

Re: Vesuvius Challenge

#13

Post by irrational » Wed, Mar 15, 2023, 8:04 PM UTC

SubiculumCode wrote: post #10

I do a lot of brain image segmentation in my research using multiatlas image segmentation, which involves diffeomorphic image registration from multiple labeled atlases...but the amount of curling in on itself of these layered sheets seems a daunting problem for a fully automated pipeline.

People spent many many decades laboriously putting tiny little dead sea scroll fragments back together like the world's worst jigsaw puzzle. I think that shows if there is a way to do this that takes a lot of tedious manual labor over many decades, there are people who will be willing to do that. They just need the tools to do the work without destroying the scrolls.

Re: Vesuvius Challenge

#14

Post by glfharris » Wed, Mar 15, 2023, 8:13 PM UTC

Fantastic project. Amazing to think that there's an entire first century library just waiting for the technology to be read.

Re: Vesuvius Challenge

#15

Post by Matt_Cutts » Wed, Mar 15, 2023, 8:15 PM UTC

For what it's worth, I worked with Dr. Seales while getting my undergrad degree. Happy to vouch that he and his team are great humans.

This is such a fascinating problem, and could have real benefits for society. Imagine uncovering ancient works that would otherwise be lost.

Re: Vesuvius Challenge

#16

Post by all2 » Wed, Mar 15, 2023, 8:24 PM UTC

janpaul123 wrote: post #2

One of the organizers here! Would love to welcome everyone's questions, ideas, etc! :)

How do I get my hands on the data set? There's no indication on the site about how to make an attempt.

Re: Vesuvius Challenge

#17

Post by all2 » Wed, Mar 15, 2023, 8:25 PM UTC

all2 wrote: post #16

janpaul123 wrote: post #2

One of the organizers here! Would love to welcome everyone's questions, ideas, etc! :)

How do I get my hands on the data set? There's no indication on the site about how to make an attempt.

Nevermind. There's a hamburger menu in the top right corner. It wasn't immediately obvious to me.

Re: Vesuvius Challenge

#18

Post by all2 » Wed, Mar 15, 2023, 9:33 PM UTC

janpaul123 wrote: post #2

One of the organizers here! Would love to welcome everyone's questions, ideas, etc! :)

Another note: the demo data [0] is behind an HTTP link. Consider getting it behind HTTPS. My browser was complaining about downloading the data from plain HTTP.

[0] https://gist.github.com/janpaul123/280262ebce904f7366fe4cc15...

Re: Vesuvius Challenge

#19

Post by mkaic » Wed, Mar 15, 2023, 9:39 PM UTC

Man, there's probably a dozen different loosely-formed ideas floating in my head right now — kudos on the exciting presentation, you really make the problem seem interesting. I may just have to give this a shot, though I think the odds of me figuring it out are exceptionally low. Still, working on cutting-edge problems is motivating, even if they're above my pay grade :)

You've nerd-sniped me good and proper. May the best team win!

Re: Vesuvius Challenge

#20

Post by janpaul123 » Wed, Mar 15, 2023, 9:55 PM UTC

Matt_Cutts wrote: post #15

For what it's worth, I worked with Dr. Seales while getting my undergrad degree. Happy to vouch that he and his team are great humans. This is such a fascinating problem, and could have real benefits for society. Imagine uncovering ancient works that would otherwise be lost.

Awesome!! They're so great; it's been such a joy to be welcomed into their project.

And totally — especially if there is indeed a larger library still waiting to be excavated. Who knows how many completely unique texts are waiting to be read, if we just have the right method to do so?!