My favorite (in terms of dark humor, if we’re honest) is YOLO, one of the more interesting deep learning object detectors. [1] It is the exact opposite of yours in every way. The code is brilliant however.
Even the papers are snarky. [2]
Even the papers are snarky. [2]
> Well the other people heavily funding vision research are the military and they’ve never done anything horrible like killing lots of people with new technology oh wait.....
> The author is funded by the Office of Naval Research and Google.
He's being funded for this. This is fantastic.
https://pjreddie.com/darknet/yolo/
https://pjreddie.com/static/Redmon%20Resume.pdf
This is the kind of guy I'd love to hang out with.
Ultimately, it boils down to people's gut feelings, which is so disgustingly random.
At a certain level, it's boring to just be good. The interesting challenge becomes to stay good while being silly as fuck.
And that's why I'm serious and professional.
> I have a lot of hope that most of the people using computer vision are just doing happy, good stuff with it, like [...] tracking their cat as it wanders around their house
And then suddenly realized I've wanted to do exactly that for a long time. Well... specifically install a camera that can detect my cat on the counter (and not my hands doing stuff) and sound an alarm/puff air to get him off.
Could this work for that? I think it could! I know my winter project...
[1] http://quinndunki.com/blondihacks/?p=3023 [2] https://news.ycombinator.com/item?id=13230904
After many months, my cat learned to stand on top of the lid and use both paws to rotate the switch forward until the lid popped open.
This blew my mind. The switch was separated from the lid. I could imagine the cat attacking the lid itself, but this separate mechanism, requiring a motion completely distinct from the motion of an opening lid, and requiring patience without instantaneous reward or even evidence that it was going to work.... I just couldn't believe it.
One example: https://sep.yimg.com/ay/entirelypets/kyjen-dog-games-slo-bow...
I love seeing a section like this when reviewing a paper. I really wish more authors would include one. (Goodness knows I've chased down enough dead-ends in some of my own research efforts.)
What's funny, though, the paper (written in a pretty much the same "fuck you" manner) is much more readable and informative than the average. Which says a lot about science papers out there.