go_away

Author Topic: Facial Recognition (Lecture)  (Read 5468 times)

0 Members and 1 Guest are viewing this topic.

Offline AdminTopic starter

  • Administrator
  • Supreme Robot
  • *****
  • Posts: 11,632
  • Helpful? 169
    • Society of Robots
Facial Recognition (Lecture)
« on: December 10, 2006, 08:41:01 PM »
Im currently working on the third tutorial in my computer vision tutorial series, so what better time than to offer a supplementary?

A lecture on facial recognition:
http://mitworld.mit.edu/video/154/
(1 hour)

"He theorizes that facial perception is a holistic process: we broadly take in the relationship, for instance, of eyes, nose and mouth. He tested this hypothesis by creating a computer program that could similarly grasp facial structure, and the program was able to “see” a face within a larger picture. In his Hirschfeld Project, Sinha is trying to distill the caricaturists’ understanding about the important landmarks of a face."

Key Points
- Vision is our most important sensor.

- Facial recognition will greatly enhance AI on mobile robots.

- "Parallel and Hierarchical Model of Vision Processing" - dividing up different vision tasks as entirely seperate

- edge detector neurons

- action potentials, signals from neurons, tuned to seeing different things

- object recognition, very fundamental problem in neuroscience

- over 1200 studies done in facial recognition in the last two decades

- image quality is important for facial recognition

- spatial configuration important (location of eyes, nose, mouth, etc.)

- facial recognition reaction time increases with lower quality images

- removing facial image degradation based on training techniques

- which aspects of facial recognition are important, and which are not?

- characterture artists 'understand facial recognition intuitively'

- facial recognition and object recognition are very similar tasks

- how does the mind collect and store object recognition information?

Offline dunk

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 1,086
  • Helpful? 21
Re: Facial Recognition (Lecture)
« Reply #1 on: December 11, 2006, 04:46:26 AM »
for anyone who is interested in Facial Recognition (or any other software based image processing) you should definitely look at this library:
http://www.intel.com/technology/computing/opencv/index.htm

they have a sample application that will pick faces in realtime out of a live video feed.
smile for the camera!

dunk.

Offline JesseWelling

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 707
  • Helpful? 0
  • Only You Can Build A Robot!
Re: Facial Recognition (Lecture)
« Reply #2 on: December 11, 2006, 09:31:23 AM »
 man it's just like an aim bot in counter strike...  :D

Offline Cognaut

  • Full Member
  • ***
  • Posts: 58
  • Helpful? 0
  • How do you know that I'm not a robot?
Re: Facial Recognition (Lecture)
« Reply #3 on: December 11, 2006, 01:16:26 PM »
Interesting lecture.  It stimulated my process.

One thing that I wonder is that, in today's technology, "blur" is most commonly recognized as a product of insufficient resolution.  We recognize the pixelation.  It's our perception though, that our eyes are not pixelated.

In the continuous world, we encounter types of bluring that have no recognizable graduations.  Like when staring at the computer screen too long, when the windows are fogged up or when spear fishing.

We have the ability to detect and approximate visual distortions in order to see through them.  I think the key to this is that we learn to simulate the distortion and use it for comparitive reference when trying to infer what's beyond it.  Not mentioned in the lecture is the role of stereoscopy and motion in the process.

Imagine a smart camera that produced 3D vector files instead of jpgs.  The files could be rendered with any resolution desired and viewed in stereo.
« Last Edit: December 11, 2006, 01:33:12 PM by Cognaut »

Offline JonHylands

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 562
  • Helpful? 3
  • Robot Builder/ Software Developer
Re: Facial Recognition (Lecture)
« Reply #4 on: December 11, 2006, 01:32:14 PM »
Imagine a smart camera that produced 3D vector files instead of jpgs.  The files could be rendered with any resolution desired and viewed in stereo.

You've just described the key to my vision system design - the real trick is figuring out a way to do this in real time (which I haven't lyet)...

- Jon

Offline Cognaut

  • Full Member
  • ***
  • Posts: 58
  • Helpful? 0
  • How do you know that I'm not a robot?
Re: Facial Recognition (Lecture)
« Reply #5 on: December 11, 2006, 01:35:49 PM »
You and I are on the same page with most of this stuff.  In the end, our models will be very similar.  I've documented my development, so you can be sure that I'm not copying.  hehe.

Maybe we should just collaborate - then again, could be stand it?  hehe.

Offline AdminTopic starter

  • Administrator
  • Supreme Robot
  • *****
  • Posts: 11,632
  • Helpful? 169
    • Society of Robots
Re: Facial Recognition (Lecture)
« Reply #6 on: December 11, 2006, 03:19:06 PM »
if you guys can do that (the 3D vector files), i want in  ;)
i see a lot of money in this . . .

Offline JonHylands

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 562
  • Helpful? 3
  • Robot Builder/ Software Developer
Re: Facial Recognition (Lecture)
« Reply #7 on: December 11, 2006, 03:26:29 PM »
Never mind 3D vector, I'd be happy with 2D vector from a camera image. It easy to do, but hard to do fast...

- Jon

Offline JonHylands

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 562
  • Helpful? 3
  • Robot Builder/ Software Developer
Re: Facial Recognition (Lecture)
« Reply #8 on: December 12, 2006, 08:57:09 AM »
I watched the lecture last night, and it was very interesting. What I got out of this lecture is that face recognition is a vector process, having mainly to do with with specific relative proportions combined with color blob analysis (shape and color of hair).

- Jon

Offline hgordon

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 373
  • Helpful? 7
Re: Facial Recognition (Lecture)
« Reply #9 on: February 03, 2007, 08:45:31 PM »
Thanks for flagging this - his method of rank ordering the attribute ratios (hair area / face area, head width / eye spacing, head width / mouth width, etc) makes a lot of sense.  It seems like you could feed the ratios into a self-organizing map or neural network with pretty good results.
Surveyor Corporation
  www.surveyor.com

 


Get Your Ad Here