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

Face detection – An overview and comparison of different solutions

liip.ch

Post reply on HN
Print view View original

11–18 of 18 posts

Re: Face detection – An overview and comparison of different solutions

#11

Post by Barrin92 »

pintxo wrote: post #2

Nice overview. Too bad the test set was rather tiny (33 images). Detection rates in this test: Amazon 52.66 % Google 40.43 % IBM 39.36 % Microsoft 17.55 %

just for comparison, I used this open source library (https://github.com/ageitgey/face_recognition) on the dataset and even with the cpu model (according to the github page the CNN is more accurate, but I couldn't test it out right now) it identified 59/188 faces.

I'm wondering what's up with the Microsoft result. Some sort of error in processing the images?

Re: Face detection – An overview and comparison of different solutions

#12

Post by bsenftner »

OkGoDoIt wrote: post #10
bsenftner wrote: post #7

Earlier quoted context omitted.

It's not open source, but our FR software runs on as low as a $100 Intel Compute Stick, while still providing 10+million FR compares per second, at accuracy rates in the 90's. On a 3.4Ghz i7 we provide 24M compares per second. That could be video or stills, acquired from live cameras, media files, or still images, via local/network storage, rtsp and related ip streams, or a REST API. Exports have full customization,…

I don't see pricing or a download/demo link on your website. "Please contact us for a facial recognition demo or to discuss your requirements for a facial recognition technology solution." My assumption from that is the price is way too high for me as a hobbyist developer, and it's not worth the hour or more of back-and-forth to try a demo and get prices. Give me clear pricing on the website and some way to access a…

No, the price is not high at all. FR tends to attract laypeople as well as sophisticated developers such as yourself. We've found that just giving a download leads to too many "casual support requests" from not real users. So, if you can stomach the stress of a human voice, you can get our software in minutes. This simple screening blocks those that would just waste our time.

Re: Face detection – An overview and comparison of different solutions

#13

Post by AndrewKemendo »

Before we compare the different face detection API’s, let's scan the images first by ourselves! How many faces would a human be able to detect?

This is great! Honestly, most researchers need to start doing this step. Baselining around accessible human capabilities (even if rough) is a super great way to show the benefits or drawbacks of using ML, especially in image processing applications, where it's more directly comparable.

Re: Face detection – An overview and comparison of different solutions

#14

Post by HEHENE »

bsenftner wrote: post #12
OkGoDoIt wrote: post #10

Earlier quoted context omitted.

I don't see pricing or a download/demo link on your website. "Please contact us for a facial recognition demo or to discuss your requirements for a facial recognition technology solution." My assumption from that is the price is way too high for me as a hobbyist developer, and it's not worth the hour or more of back-and-forth to try a demo and get prices. Give me clear pricing on the website and some way to access a…

No, the price is not high at all. FR tends to attract laypeople as well as sophisticated developers such as yourself. We've found that just giving a download leads to too many "casual support requests" from not real users. So, if you can stomach the stress of a human voice, you can get our software in minutes. This simple screening blocks those that would just waste our time.

I'm actively working on facial recognition in a retail store context, but I also quickly dismissed your product due to a lack of pricing information.

3D-AI looks like it would be a great fit, but "Call for demo" translates to "probably $100,000+ with lengthy on-boarding and extortion-level support contracts"

I don't mind calling or emailing to get a trial license, but I'm not going to waste my time or yours if I don't even know if it'll be in-budget.

Re: Face detection – An overview and comparison of different solutions

#15

Post by symisc_devel »

HEHENE wrote: post #14
bsenftner wrote: post #12

Earlier quoted context omitted.

No, the price is not high at all. FR tends to attract laypeople as well as sophisticated developers such as yourself. We've found that just giving a download leads to too many "casual support requests" from not real users. So, if you can stomach the stress of a human voice, you can get our software in minutes. This simple screening blocks those that would just waste our time.

I'm actively working on facial recognition in a retail store context, but I also quickly dismissed your product due to a lack of pricing information. 3D-AI looks like it would be a great fit, but "Call for demo" translates to "probably $100,000+ with lengthy on-boarding and extortion-level support contracts" I don't mind calling or emailing to get a trial license, but I'm not going to waste my time or yours if I don'…

You may take a look to SOD embedded project which is open source and do support facial detection at Real-time.

https://github.com/symisc/sod

Re: Face detection – An overview and comparison of different solutions

#16

Post by bsenftner »

HEHENE wrote: post #14
bsenftner wrote: post #12

Earlier quoted context omitted.

No, the price is not high at all. FR tends to attract laypeople as well as sophisticated developers such as yourself. We've found that just giving a download leads to too many "casual support requests" from not real users. So, if you can stomach the stress of a human voice, you can get our software in minutes. This simple screening blocks those that would just waste our time.

I'm actively working on facial recognition in a retail store context, but I also quickly dismissed your product due to a lack of pricing information. 3D-AI looks like it would be a great fit, but "Call for demo" translates to "probably $100,000+ with lengthy on-boarding and extortion-level support contracts" I don't mind calling or emailing to get a trial license, but I'm not going to waste my time or yours if I don'…

Actually, try $2000 and you own a license. Support is negotiated, and if you're technical nearly nothing. Support is typically for the non-technologists. None of the major players in FR list their prices on the sites, BTW. Just look them up at the NIST FR Challenge website; you'll see us in the top ten year after year, and you won't see anyone from the article above.

Oh of, 3D-AI: I'm the lead.

Re: Face detection – An overview and comparison of different solutions

#17

Post by kaivi »

symisc_devel wrote: post #9
kaivi wrote: post #8

I've tinkered with face recognition in the past days, and am now waiting for a response to my GCE quota request for a V100 GPU. $8/h rate for a preemptible instance seems really cheap, yet I am not sure if I will ever be able to process my dataset. Does anyone have a clue on how much it will cost to detect faces and extract 128d encodings from ~100M of 200x200 photos?

PixLab do offer quota based monthly plans for such task including standard face detection, shape extraction, gender, age and emotion pattern extraction. They do charge $0.9 per 1000 requests after you reach your monthly quota (1.1M API calls). https://pixlab.io/cmd?id=facedetect .

Woah, that would be $80k to process my dataset. I'll try renting hardware instead.

Re: Face detection – An overview and comparison of different solutions

#18

Post by symisc_devel »

kaivi wrote: post #17
symisc_devel wrote: post #9

Earlier quoted context omitted.

PixLab do offer quota based monthly plans for such task including standard face detection, shape extraction, gender, age and emotion pattern extraction. They do charge $0.9 per 1000 requests after you reach your monthly quota (1.1M API calls). https://pixlab.io/cmd?id=facedetect .

Woah, that would be $80k to process my dataset. I'll try renting hardware instead.

Not that much. Contact them and they will bring back with an offer adequate to your dataset and budget.
Post reply on HN