Joseph Redmon

While I don't disagree with you, I think we can have better conversations on HN. Honestly I would love to hear from someone with a different opinion than me. I really do want to hear the counter argument. But this is also the technology that made Joseph Redmon[0] (creator of YOLO) leave vision research. We can have ethical discussions. We should have ethical discussions. And different opinions are how we learn. And I do think HN is one of those places to have these ethical conversations. [0] https://pjreddie.com/

> Why is furry/anime/whatever inherently disqualifying, regardless of the quality of the technical content? It isn't inherently disqualifying, unless it is plastered everywhere where it doesn't belong, like in the middle of a lot of unrelated conversations or in the middle of a professional technical blog post. For example, if you just have a furry image in your website header, but the content of your technical post itself is on-point and doesn't have a bunch of unrelated furry stuff, most people will have no issues sharing your content and recommending it to others. For a good example of that, take a look at the website of the guy who did a lot of impressive work and research and became famous for the YOLO image classifier[0]. His website features my little pony characters. He is very openly into it. His MLP-themed resume made as a halfway joke is extremely infamous on the internet. And then take a look at his technical blog posts. He writes really well and doesn't let his interests detract him from quality writing. And no one who is interested in reading about the technical topics he covers seems to have any issues with the content of his posts at all. 0. https://pjreddie.com/

One of my favorite examples from that era: https://pjreddie.com/static/Redmon%20Resume.pdf And in case you were wondering what this little pony did next... https://scholar.google.com/citations?user=TDk_NfkAAAAJ&hl=en

> Far too few people in CS care about the ethics side as well. Unrelated to this story, but on the topic of ethics in CS: I found the decision of Joe Redmon [1] extremely brave. (If you don't know who he is, he's the principle original author of YOLO, and could easily have had a stellar CV career. Also, his papers are just amazing reads with tons of humor.) [1] https://syncedreview.com/2020/02/24/yolo-creator-says-he-stopped-cv-re… [2] https://pjreddie.com/darknet/yolo/

the YoloV3 (the model that these tools are designed to work with) paper is extremely funny and worth reading for anyone who hasn't. https://pjreddie.com/media/files/papers/YOLOv3.pdf

> But maybe a better question is: “What are we going to do with these detectors now that we have them?” A lot of the people doing this research are at Google and Facebook.I guess at least we know the technology is in good hands and definitely won’t be used to harvest your personal information and sell it to.... wait, you’re saying that’s exactly what it will be used for?? Oh. > 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.

Is there a more self-contained way of running trained models on devices like the Pi? Almost all tutorials I've come across always require me to install a lot of Python dependencies. I'd love to have classifiers installed as a single binary somewhere. I guess https://pjreddie.com/darknet/yolo/ is pretty close to that and I would imagine with a custom build wrapper around https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contri… the same could be achieved. Any hints?

Maybe https://pjreddie.com/darknet/install or one of the various forks. YOLO is supposed to be one of the best object detection systems as far as accuracy/speed is concerned.

This article inspired me to have a play around with Darknet and Darkflow - turns out they're pretty easy to get going on an OS X laptop with Python 3 (installed via Homebrew). Here's how I got Darkflow working: https://gist.github.com/simonw/0f93bec220be9cf8250533b603bf6dba For Darknet, I just ran "make" as documented here: https://pjreddie.com/darknet/install/ and then followed the instructions on https://pjreddie.com/darknet/yolo/ and https://pjreddie.com/darknet/nightmare/ to try it out.

The biggest difference between our two projects, is the effort you've put into training your own model, which I think is amazing. One of the technical issues that you pointed out, is that a model trained on still images, shouldn't be expected to work on video. While I did not train a custom model for this this project, I'm current working on another DNN model for a completely different purpose, where I think feeding frame deltas into the model, will improve the outcome. As a hobbyist, I would reckon for porn and the like, analysing frozen frames is probably just enough. For violence however, I would agree with you and say that some effort to encode motion would be essential. Focussing on NSFW content generally, I would guess, depending on the scale of your project, that you will forever run into 'edge cases' for NSFW images, even before you run into the soft wall of subjectivity. I agree that the tech is improving all the time, and I think something like this can be made to be truly useful one day. Possibly soon. But it would need a large, active development team, a great deal more compute, and a LOT of data. In much the same way that no home/garage coder can hope to put together a model like GPT3 right now, I would think that a foolproof NSFW classifier would need more resources than you or I have access to at this moment. But things change all the time. Thinking about what you're doing, one thing I might suggest, if you have time to develop it, is to add some kind of…

Also, best cv: https://pjreddie.com/resume/

I think 200k as an average is good, since some will be making a whole lot more, and there will be the 100K per year positions too. My equivalent to your dearth of people in the know in 2011, was studying AL (Artificial Life) in 1990s, and nobody heard of it. It included the study of ANNs, GAs, GP and AI in general (which I prefer to call CI - Computational Intelligence nowadays). The book that started it all for me [0]. There were no immediate applications aside from expert systems here and there, and the fuzzy logic appliance controllers coming out of Japan. However, now, with self-driving cars, recommender systems, image recognition (face-matching surveillance post 9/11) have big pay-offs or budgets to spend on it. VR is having its second renaissance. I thought the first time it would have taken off even with the clunky glasses and headsets, since gaming was already such a huge money industry. And now with modeling becoming prominent again, AL paradigms are being modified, created and repurposed for all sorts of cool things. I play with NetLogo since it is a fun environment for that [1]. My only regret is that I was in at the beginning, but left it to pursue other things, and so I am not at the level of practice to get one of those 200K jobs. I still kept studying it all these years though, and I have coded my own bits and pieces, but mainly for conceptual pieces, art and music, not practical applications. I play with Darknet [2] now, since C was my second language…

I think it's intresting to read about the guy who made yolo. Take a look at his website and later, his thoughts about the monster he may have created. https://x.com/pjreddie/status/1230524770350817280 https://pjreddie.com/darknet/yolo/ https://medium.com/@graham.wallington/the-evolution-of-yolo-joseph-red…

Yes in technical sense that only wrong countries try to control wrongthink, and no in the sense that if there are chances such opinions prevail. It'll be like insisting YOLO[1] renamed and presentation video changed up. We all kind of deserve a right to say so, and insisting anime is unprofessional and must be removed don't look that way yet, but IMO it very well could be considered insane talks in 5-10 years. 1: https://pjreddie.com/darknet/yolo/

How does CLIP compare to YOLO[1]? I haven't looked into image classification/object recognition for a while, but I remember that YOLO was quite good was working on realtime video too. [1]: https://pjreddie.com/darknet/yolo/

There's also the funding acknowledgement in the YOLO v3 tech report (https://pjreddie.com/media/files/papers/YOLOv3.pdf): > But maybe a better question is: “What are we going to do with these detectors now that we have them?” A lot of the people doing this research are at Google and Facebook. I guess at least we know the technology is in good hands and definitely won’t be used to harvest your personal information and sell it to.... wait, you’re saying that’s exactly what it will be used for?? Oh. > 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....[1] > [1] The author is funded by the Office of Naval Research and Google.

OpenCV and "AI" can work well together; see YOLO: https://pjreddie.com/darknet/yolo/

And miss out on gems like this? https://pjreddie.com/static/Redmon%20Resume.pdf

Content is solid. I'm afraid I can't ignore the author's resume link... https://pjreddie.com/static/Redmon%20Resume.pdf

The charts in this paper are hilarious: https://pjreddie.com/media/files/papers/YOLOv3.pdf Previous authors didn’t start their axes at 0, so he kept their axes and just put the timing for YOLO outside the original chart area.

YOLOv3 is a pretty delightful read https://pjreddie.com/media/files/papers/YOLOv3.pdf

https://pjreddie.com/darknet/yolo/ Is that pjreddie used horses, dogs, and bicycles as training data? Not realising that his technology could also be used on human faces?

For those not familiar with the guy, here's his CV (Curriculum Vitae, not Computer Vision this time): https://pjreddie.com/static/Redmon%20Resume.pdf

for those who aren't familiar with the author, he previously worked at metamind / salesforce research doing nlp and has published many successful nlp papers [0]. he opted to write an informal paper for this project (similar to yolov3 [1]), but the work itself should still be taken seriously. [0] https://scholar.google.com/citations?user=AolIi4QAAAAJ [1] https://pjreddie.com/media/files/papers/YOLOv3.pdf

https://pjreddie.com/media/files/papers/YOLOv3.pdf Sounds to good to be true. Also reads like that. :) A gem from this paper: But maybe a better question is: “What are we going todo with these detectors now that we have them?” A lot ofthe people doing this research are at Google and Facebook.I guess at least we know the technology is in good handsand definitely won’t be used to harvest your personal infor-mation and sell it to.... wait, you’re saying that’s exactlywhat it will be used for?? Oh.Well the other people heavily funding vision research arethe military and they’ve never done anything horrible likekilling lots of people with new technology oh wait..... 1 1 The author is funded by the Office of Naval Research and Google.

Nothing beats this resume https://pjreddie.com/static/Redmon%20Resume.pdf

Try Yolo: https://pjreddie.com/darknet/yolo/ . It's written in C.

I would not dismiss C or C++ as viable avenues of progress. There are viable C++ / C only deep learning libraries https://pjreddie.com/darknet/, available that are easily ported and embedded C++ can be pretty pleasant to write. We currently use C++14 on a embedded Power processor with custom vector unit and 16kB of SRAM and it works like a charm.

Looking at the resume, I noticed I've seen his resume shared on HN before as an example of an unconventional looking resume. Anybody remember the resume with the ponies? https://pjreddie.com/static/Redmon%20Resume.pdf

You should look up the guy who built YOLO (an object detector with 1700 citations). His papers: https://pjreddie.com/media/files/papers/YOLOv3.pdf and his resume: https://pjreddie.com/static/Redmon%20Resume.pdf are all written in this style.