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

DeepFace: Closing the Gap to Human-Level Performance in Face Verification

facebook.com

Post reply on HN
Print view View original

41–46 of 46 posts

Re: DeepFace: Closing the Gap to Human-Level Performance in Face Verification

#41

Post by Pwnguinz »

bayesianhorse wrote: post #34
userbinator wrote: post #23

Earlier quoted context omitted.

> For one thing, this isn't exactly a stealthy or cheap thing to do. That's actually the more disturbing part about this: it require computational resources that only governments and large corporations can afford, so they are the ones who gain the most from it; and it gives them more leverage over the population.

"Leverage" is kind of vague. It's hard to make money from large-scale abuses... And it doesn't look like the NSA for example is doing anything useful from its own perspective with the data they actually have...

"Leverage" doesn't always equate to profit (or even revenue). The "goal" of the NSA is not profit, but mass surveillance. Whereas surveillance historically has been centralized, with the onslaught of mobile devices, we now have prevalent distributed surveillance (and perhaps more so than surveillance).

All the data everyone is feeding into FB/G+/Twitter/etc about their friends and acquaintances could not have been collected on such a mass scale by the NSA alone.

If you're a privacy conscious individual, there's definitely reasons to be fearful of this approach to surveillance by any governmental intelligence agency.

Re: DeepFace: Closing the Gap to Human-Level Performance in Face Verification

#42

Post by cs702 »

apu wrote: post #37
cs702 wrote: post #17

The key innovation is an accurate, reliable method for rotating faces so they're 'looking straight at the camera' before feeding them to a deep neural network. They call this 3D photo rotation process "frontalization." Figure 1 on page 2 of the paper shows at a very high level how this is being done. Very nice!

Actually, that's only one of the contributions, and I'm not so sure it's the "key innovation". Every other recent face recognition method also tries to do some kind of alignment to make faces more similar in pose/expression/lighting prior to classifying them; and of these, several also fit faces to a 3-d model to rotate to frontal (with varying quality). See my other comment for my guess on what's actually providing…

Yes, other attempts "also fit faces to a 3-d model to rotate to frontal (with varying quality)," as you put it, but this method for rotating faces appears to be superior -- more accurate and reliable.

It looks like the main contribution to me :-)

Re: DeepFace: Closing the Gap to Human-Level Performance in Face Verification

#43

Post by boomzilla »

apu wrote: post #36

Having worked on this problem before (the comparison to human performance they cite is from my work) and seeing all the recent successes of deep learning, I'd bet that a lot of the gain here comes from what deep learning generally provides: being able to leverage huge amounts of outside data in a much higher-capacity learning model. Let me try to break this down: In machine learning, when you have input data that is…

Thanks for the write-up. This is very informational.

Could you elaborate a bit more on the "capacity" of learning models? Can it be quantified and is it some how related to the VC dimension of a particular learning problem? It would be great if you could give some example of "capacity" for the more well known models: trees, naive bayes, SVM, one hidden layer neural nets, etc.

Re: DeepFace: Closing the Gap to Human-Level Performance in Face Verification

#44

Post by apu »

boomzilla wrote: post #43
apu wrote: post #36

Having worked on this problem before (the comparison to human performance they cite is from my work) and seeing all the recent successes of deep learning, I'd bet that a lot of the gain here comes from what deep learning generally provides: being able to leverage huge amounts of outside data in a much higher-capacity learning model. Let me try to break this down: In machine learning, when you have input data that is…

Thanks for the write-up. This is very informational. Could you elaborate a bit more on the "capacity" of learning models? Can it be quantified and is it some how related to the VC dimension of a particular learning problem? It would be great if you could give some example of "capacity" for the more well known models: trees, naive bayes, SVM, one hidden layer neural nets, etc.

Yes, capacity is intimately tied to VC dimension; in particular, VC dimension is one way to measure capacity. See the Wikipedia article for more information: http://en.wikipedia.org/wiki/Vc_dimension

I'm not an expert on deep learning (although I generally understand how they work on vision problems), so I'm not sure if you can precisely measure the capacity of deep networks. Informally, the primary number that seems to matter is the number of parameters in the network that have to be learned. This paper quotes that at "more than 120 million".

SVMs, in contrast, typically work with feature dimensionalities (i.e., # of parameters) that are on the order of 1,000 - 100,000. You can't directly compare these numbers because there are various non-linearities involved, but this deep learning network is definitely much higher capacity than an SVM would be with normal feature dimensionalities.

Re: DeepFace: Closing the Gap to Human-Level Performance in Face Verification

#45

Post by somberi »

apu wrote: post #36

Having worked on this problem before (the comparison to human performance they cite is from my work) and seeing all the recent successes of deep learning, I'd bet that a lot of the gain here comes from what deep learning generally provides: being able to leverage huge amounts of outside data in a much higher-capacity learning model. Let me try to break this down: In machine learning, when you have input data that is…

I envy your ability to write clearly and combine different threads together.
Post reply on HN