article in Personal
projects-and-research
Facial recognition - Teaching the machine to recognize like we do.
Going through school you start to realize just how little we understand about everything going on in our body. Basic biology courses seem to skim over the how and focus more on name memorization and broad function. This skimming is esp. true when we study the way our minds work. How does the mind actually understand the sensory information we receive? How exactly does our memory work? How do we recognize people, places, and things?
The body and the mind is a complex machine indeed. My fascination in facial recognition started when I was in high school. During my junior and senior year, I studied artificial neural networks and their biological counterparts. I decided to create a neural network to recognize faces.
Here is my overview presentation:
[HTML] [PowerPoint]
Soon to come, my research papers from my facial recognition project in 2000.
--Dave
When I was in high school, I took part in a science project semester where I studied artificial neural networks. The idea is pretty simple. Build digital models of neurons and simulate within a computer. The question is how to engineer the neurons so that they perform some meaningful task. Connected together, using different means of "training" the network learns to differentiate inputs.
Face Detection C++ library with Skin and Motion analysis - The Code Project - C++ / MFC
Sources for test data
We should be able to train the computer to recognize faces using a set of faces which we've applied metadata to. The computer will learn to recognize the defining characteristics of the test data. Therefore, we must work hard to create a large enough training set to allow for the greatest generalization. (it'll be as good as the data we feed it)
When I started my science fair project, I used the database....
OnThaFly Users
Commercial face trackers and detectors
faceAPI
LFW Face Database : Main
Created: 2007-11-06 19:38:55
Modified: 2014-04-24 00:56:56
