Nvidia's demo of real-time object recognition using deep learning [video]
1–10 of 24 posts
Re: Nvidia's demo of real-time object recognition using deep learning [video]
#2Re: Nvidia's demo of real-time object recognition using deep learning [video]
#3Looks great! Can anyone with more knowledge about deep learning say whether this is an exceptional achievement in the field ?
http://cs.stanford.edu/people/karpathy/deepimagesent/
I think the impressive thing here is that the GPU is presumably doing GIANT matrix multiplications in real time. A prediction from a neural net is just a series of matrix multiplications, and matrix multiplications are about n^2.8 in complexity, so you can see how matrix multiplications with thousands of rows/columns (often what these sorts of deep image classifiers involve) are hugely computationally expensive.
So it's definitely important for real time machine learning systems to have access to this kind of linear algebra power, but the actual machine learning techniques demonstrated are not super impressive. The hardware is. Which makes sense since this is an Nvidia demo.
Re: Nvidia's demo of real-time object recognition using deep learning [video]
#4Looks great! Can anyone with more knowledge about deep learning say whether this is an exceptional achievement in the field ?
However, building a real world working system has challenges that are different to the academic challenge of trying to classify the most classes possible in static images.
[1] http://blogs.technet.com/b/machinelearning/archive/2014/11/1...
Re: Nvidia's demo of real-time object recognition using deep learning [video]
#5Looks great! Can anyone with more knowledge about deep learning say whether this is an exceptional achievement in the field ?
Re: Nvidia's demo of real-time object recognition using deep learning [video]
#6Looks great! Can anyone with more knowledge about deep learning say whether this is an exceptional achievement in the field ?
As far as state-of-the-art classification goes, this is not terribly impressive. See the recent results in something like: http://cs.stanford.edu/people/karpathy/deepimagesent/ I think the impressive thing here is that the GPU is presumably doing GIANT matrix multiplications in real time. A prediction from a neural net is just a series of matrix multiplications, and matrix multiplications are about n^2.8 in complexit…
Re: Nvidia's demo of real-time object recognition using deep learning [video]
#7Earlier quoted context omitted.
As far as state-of-the-art classification goes, this is not terribly impressive. See the recent results in something like: http://cs.stanford.edu/people/karpathy/deepimagesent/ I think the impressive thing here is that the GPU is presumably doing GIANT matrix multiplications in real time. A prediction from a neural net is just a series of matrix multiplications, and matrix multiplications are about n^2.8 in complexit…
From what i can understand what's even more impressive is that it was running on a beefed up version of their latest mobile SOC and not on some 5000$ compute GPU card. Which means that this application can be both very affordable and very practical since people won't put a 300W GPU in their car.
Right now you have iPhones and whatnot doing touch ID with fingerprints, but imagine if your phone could recognize you just by quickly analyzing the gyro data as you raise your phone and comparing it against the other thousand times you've pulled your phone out of your pocket, combined with the slight pressure readings near the touchscreen's edge because it's learned where you're fingers fall on the case.
^ contrived example I just thought of, but you get the idea.
Re: Nvidia's demo of real-time object recognition using deep learning [video]
#8on the right merc sls classified as SUV
on the left one SUV classified as two VANs
Their algorithm works at about 1Hz rate when doing signs. This is ~state of the art from 20 years ago, but running on small mobile SoC at a slow rate.
Re: Nvidia's demo of real-time object recognition using deep learning [video]
#9Earlier quoted context omitted.
From what i can understand what's even more impressive is that it was running on a beefed up version of their latest mobile SOC and not on some 5000$ compute GPU card. Which means that this application can be both very affordable and very practical since people won't put a 300W GPU in their car.
Definitely agreed. When well-maintained and easy to use machine learning libraries meet very powerful, highly embeddable GPUs (or other dedicated linear algebra sort of hardware), I think we'll see a big revolution in the whole "smart object" field. Right now you have iPhones and whatnot doing touch ID with fingerprints, but imagine if your phone could recognize you just by quickly analyzing the gyro data as you rais…