Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Building an AI that watches rugby

nickjones.tech

Post reply on HN
Print view View original

21–30 of 52 posts

Re: Building an AI that watches rugby

#21

Post by dmurray »

MuffinFlavored wrote: post #12
dmurray wrote: post #8

Earlier quoted context omitted.

The only interesting part of the model's output was { "current_play": "ruck", } So the vision model can correctly identify that there's a ruck going on and that the ball is most likely in the ruck. Why not build on this? Which team is in possession? Who was the ball carrier at the start of the ruck, and who tackled him? Who joined the ruck, and how quickly did they get there? How quickly did the attacking team get th…

ESPN has play by play stuff for free like this on their website for some other sports not sure if it is done by a human or not curious how “an AI can do it” yields much difference in terms of result for the casual watcher

> curious how “an AI can do it” yields much difference in terms of result for the casual watcher

An AI can do it in volume, and therefore cheaper. I don't think a human could do everything I said in real time - maybe with a lot of training and custom software.

A human could transcribe the scoreboard, but the article still thinks that's an interesting application of cutting-edge machine vision.

Re: Building an AI that watches rugby

#23

Post by ookdatnog »

brookst wrote: post #17
ookdatnog wrote: post #15

Earlier quoted context omitted.

The AI's job as described in this article is two-fold: - The relatively trivial task of extracting textual data from the screen. - The task of obfuscating that they're publishing other people's work as their own. When I clicked the article I assumed they'd try to automatically construct analysis of the game by using AI to analyze frames of the game, but that's not what they are doing. They are extracting some trivial…

Perhaps the most surprising thing about the whole LLM revolution is how quickly attitudes about IP have shifted in the HN and similar communities. A few years ago, media companies were rent-seeking parasites who leveraged the jack-booted thugs of law enforcement to protect an artificial monopoly using IP laws that were massive overreach and contrary to the interests of humanity. Today, suddenly, media companies are p…

Not commenting on general trends, but I don't think my opinion on IP shifted massively as a result of the rise of LLMs. I can summarize it as follows:

- It seems desirable to have some system that allows creatives to be paid for their work.

- Whether current IP law is the best system we can come up with is highly debatable. But nevertheless it is the system we have, and its existence is to some extent justified.

- If we look at the "pefect case" where IP law functions as intended (for example, an author publishes a book in which they invested years of their life), then breaking IP law (sharing that author's work without their consent) in that instance seems, to me, immoral.

- Nevertheless there are plenty of excesses in the system where I would judge that the application of IP law is unjustified and breaking the law is morally justified (naturally I still don't recommend it). This includes, for example, paywalled papers from publicly-funded research, works that can no longer reasonably be purchased (for example games for old consoles), most if not all software patents, ...

So the question simply boils down to: is sports commentary justifiably protected under IP law? I think the answer is a pretty clear-cut "yes" here, I don't see how it falls under any case of IP law overreach.

Re: Building an AI that watches rugby

#24

Post by sebastiennight »

I love that as soon as he writes,

> The plan was simple.

You know you're in for a funny read.

More seriously though, the JSON example from a vision language model is interesting but does not take into account how much extrapolation (hallucination) the model will insert over time.

For instance, even if not visible in the image, your VLM will probably start inserting details (such as the color of the team's jersey) based on knowing the team's three-letter identifier.

So the reliability of the system will go down over time, and it probably compounds if you're using some of that info to feed further steps in the loop.

Re: Building an AI that watches rugby

#25

Post by sebastiennight »

djtango wrote: post #4

I don't quite get how diffing frames allows you to find the scores. TFA mentions comparing a frame with and without - but how do you generate that frame without? If you can already do it, what's useful about doing that?

He's diffing the frames, and then the only pixels that stay the same are the UI, from which he doesn't directly get the UI (see the example, it's illegible) but he can extract the POSITION of the UI on the screen by finding all the non-red pixels.

And then he does a good ol' regular crop on the original image to get the UI excerpt to feed the vision model.

Re: Building an AI that watches rugby

#26

Post by hummuscience »

The moment I started reading this, I got reminded of this recent study: https://arxiv.org/html/2503.10212v1

The scope is a bit different. The study uses an LLM to interpret pose estimation data and describe the behavior in each frame. The output is text which can be used to create embeddings of behavior. As someone who works in ethology, that's a clever (but maybe expensive) idea.

I think the author could use something similar. With multi-person pose estimation models.

Re: Building an AI that watches rugby

#27

Post by nirvael »

barbegal wrote: post #13
djtango wrote: post #4

I don't quite get how diffing frames allows you to find the scores. TFA mentions comparing a frame with and without - but how do you generate that frame without? If you can already do it, what's useful about doing that?

I think the text is wrong, it's diffing two frames and the areas that are the same are where the scorebaord is as this doesn't change between frames but everything else does.

I was also confused by this. I think you're right, but in the original text they specifically mention a 'static background' that they remove, so it's not just a simple 'wrong way round' error, it's a fundamental misunderstanding of what's happening. Makes me wonder if the author actually knew what they were doing, or just using an LLM to vibe-code everything.

Re: Building an AI that watches rugby

#28

Post by Workaccount2 »

bondarchuk wrote: post #19
brookst wrote: post #17

Earlier quoted context omitted.

Perhaps the most surprising thing about the whole LLM revolution is how quickly attitudes about IP have shifted in the HN and similar communities. A few years ago, media companies were rent-seeking parasites who leveraged the jack-booted thugs of law enforcement to protect an artificial monopoly using IP laws that were massive overreach and contrary to the interests of humanity. Today, suddenly, media companies are p…

How do you know you're seeing peoples' opinions change, and not just a change in which people express their opinions? That said I'd personally be happy if LLMs cause the death (or drastic weakening) of copyright and IP laws, however as it is now, with no copyright for AIs but the same old copyright for humans, it's the worst of both worlds.

I know people personally with strong gripes about AI "infringement" (in quotes because I believe people are just confused about how these models work), and every single one of them -100%- have a stash of pirated media they casually accumulated over the years.

People are in it for themselves. When you are young everyone has righteous ideals, but then trends of society eventually ebb, and you realize that just about everyone was simply virtue signalling, and few people are committed even to their own detriment.

2005: "End copyright! Trash IP law! Liberate media!"

2025: "Strengthen Copyright! Extend IP Protection! Protect makers!"

Post reply on HN