Here is the face detection demo I ran: http://liuliu.me/detect/detect.html#http%3A%2F%2Fimg.dailyma...

(press the Detect button when the page loads)

That is impressive.

The blog post is a little light on details. Just go to the source and demo on the original creator's site: http://liuliu.me/detect/detect.html

Canvas allows access to individual pixels in an image, and JS is arguably faster than BASIC interpreters of the 80s. Fun things lie ahead in client-side image processing.

EDIT: Incidentally, the code is pretty hairy and has about 3 comments. It's hard to figure out what ML approach he's using from a cursory look.

EDIT: Aah, definitely an image pyramid in there, so some multi-resolution stuff happening.

http://squidder.com/face/ - real time face detection with flash

It seems to on average get about half the faces. http://liuliu.me/detect/detect.html#http%3A%2F%2Fwww.alexrei... http://liuliu.me/detect/detect.html#http%3A%2F%2Fkk.org%2Fct... It recognizes upside down faces though, which is impressive. http://liuliu.me/detect/detect.html#http%3A%2F%2Fwww.scifun.... Overall, a great demo.

I did this one: http://liuliu.me/detect/detect.html#http://antikewl.com/dail...

And it only detected Jobs, but not Woz ...

Trouble with false positives : http://liuliu.me/detect/detect.html#http%3A%2F%2Ffarm1.stati...

Some of the bricks look like the Face on Mars :) http://bit.ly/aWTgzA

I don't know too much about face detection, can you anyone tell which algorithm the author is using? It doesn't work on a lot of pictures so I was wondering if this is the best face detection algorithm available?

I didn't look at the source in detail, but almost all face detection systems today (including those implemented in hardware in cameras) use a variant of Viola and Jones' Haar-Like Cascade Detector [1].

The major exception is work from CMU and PittPatt [2], which uses Neural Networks [3].

[1] http://en.wikipedia.org/wiki/Viola-Jones_object_detection_fr...

[2] http://www.pittpatt.com/

[3] http://vasc.ri.cmu.edu/NNFaceDetector/

I doubt that PittPatt uses NN. Here is the reference I found:

Feature-Centric Evaluation for Efficient Cascaded Object Detection, Henry Schneiderman

Maybe Schneiderman himself can clarify this :)

Ah good point. I forgot that he had his own method, separate from Takeo's.

Hi, I put an introduction to the particular implementation behind it:

https://github.com/liuliu/ccv/blob/current/doc/bbf.md

You can checkout the first paper, it is a light reading.

Doesn't really seem to work that good on pictures with more objects on it.

There already is a Facebook app to aide in tagging photos by detecting faces:

http://apps.facebook.com/photo-finder/

It's actually quite accurate too...

It's pretty sweet. I wrote qoobster.com/bieber to get Antoine Dodson's sweet headband instead of that silly red square plastered on my friends.

haha, nice!