This is great! I do wish this were written in something other than Python. What is the carbon footprint of all this computer vision, compute-intensive code still being run billions of times a day in Python? Someone should calculate...
Facebook open-sources Detectron
11–20 of 188 posts
Re: Facebook open-sources Detectron
#12So is this the end of Google captchas asking for where the car/sign/whatever is? Will there be a final battle of AIs, where they will kill each other, and the unfettered access to websites over VPN/tor wins and laughs the last laugh?
This is some of the most advanced work out there - but CV is not “solved” most vision systems only can label about 1k categories of objects. So capatchas can still be easiy constructed that would fool these systems. Part of why it is exciting to get this out there others can help us improve it.
Imagenet is the only reason why most models are trained on 1K of categories. There's plenty of models in the wild that handle 10s of thousands of classes.
Re: Facebook open-sources Detectron
#13This is great! I do wish this were written in something other than Python. What is the carbon footprint of all this computer vision, compute-intensive code still being run billions of times a day in Python? Someone should calculate...
The actual computationally-hard part of the code is run in the GPU using CUDA.
Re: Facebook open-sources Detectron
#14I'm happy that tech companies are open sourcing basic research all the time, and thinking a lot about what would have happened if large pharmacy companies did the same thing. I'm just hopeful that with new biotech companies the science behind curing people will get faster as well.
Unlike the case in tech, pharma basic research is far less important in advancing our knowledge when compared to academia. A good example comes from the last few blockbuster cancer therapies - CAR-T cells and checkpoint blockade all arose in academic labs.
Also, for drugs that do make it to market, efficacy and side effect information is published as a condition of drug approval, at least for new drugs.
Whether basic science research papers should be behind a paywall is a wholly separate issue, but the life science community largely shares its finished products. Indeed, there’s even a push to share early stage data, too.
Re: Facebook open-sources Detectron
#15Earlier quoted context omitted.
This is some of the most advanced work out there - but CV is not “solved” most vision systems only can label about 1k categories of objects. So capatchas can still be easiy constructed that would fool these systems. Part of why it is exciting to get this out there others can help us improve it.
Imagenet is the only reason why most models are trained on 1K of categories. There's plenty of models in the wild that handle 10s of thousands of classes.
With what precision and recall?
Re: Facebook open-sources Detectron
#16> Beyond research, a number of Facebook teams use this platform to train custom models for a variety of applications including augmented reality and community integrity. Any idea what they mean by "community integrity"?
detecting porn, presumably.
I would expect it to work on detecting the mismatching of content for types of communities in general. For example, preventing pictures of cats or giraffes being uploaded as a product photo on Poshmark when it's supposed to be a pair of shoes.
That type of check should become standard in a short amount of time for all communities that accept photos (that isn't meant to be general purpose, eg Imgur).
Re: Facebook open-sources Detectron
#17Earlier quoted context omitted.
Imagenet is the only reason why most models are trained on 1K of categories. There's plenty of models in the wild that handle 10s of thousands of classes.
With what precision and recall?
Not that much worse than what you see on imagenet. Most large companies have internal datasets with >100 million images.
Re: Facebook open-sources Detectron
#18Earlier quoted context omitted.
detecting porn, presumably.
I would expect it to work on detecting the mismatching of content for types of communities in general. For example, preventing pictures of cats or giraffes being uploaded as a product photo on Poshmark when it's supposed to be a pair of shoes. That type of check should become standard in a short amount of time for all communities that accept photos (that isn't meant to be general purpose, eg Imgur).
A few craigslist competitors are using CV to detect drugs/weapons/animals.
Re: Facebook open-sources Detectron
#19Earlier quoted context omitted.
detecting porn, presumably.
I would expect it to work on detecting the mismatching of content for types of communities in general. For example, preventing pictures of cats or giraffes being uploaded as a product photo on Poshmark when it's supposed to be a pair of shoes. That type of check should become standard in a short amount of time for all communities that accept photos (that isn't meant to be general purpose, eg Imgur).
For example marketplace (where you can sell items on Facebook) will suggest a category for the item if you upload a photo.
Re: Facebook open-sources Detectron
#20This is great! I do wish this were written in something other than Python. What is the carbon footprint of all this computer vision, compute-intensive code still being run billions of times a day in Python? Someone should calculate...
What was the carbon footprint of the turk machines the Python can replace?