Hey I am looking to design an algorithm in Matlab which is able to highlight faces in an image. I have posted this in the programming forum but not many people post there and so I have had little success. Basically I want to use an approach known as Eigenfaces, I am a little stuck and was wondering if anyone here has any experience?
face detection using matlab
- Thread starter SoSolid
- Start date
Another thread?
If you want to know more about Eigenfaces, check out the papers at the bottom of the Wiki entry, especially Turk & Pentland.
http://en.wikipedia.org/wiki/Eigenface
hehe. I have read through lots of papers etc but the problem is I am not experienced in this area and basically don't understand anything in the papers. I want to see examples of actual working code so I can understand how it works. I get lost at the shear amount of mathematics involved in some of those papers.
Do your own homework
It is not homework and I am doing it on my own, now sit down.
Difficult in that a final year project could be drawn from it at least, yes
Yeah I know lol I am only in my second year so this is a real meaty project.
Can you send me som info or help me in any way, I forgot to mention I am doing a degree in Mechanical Engineering so have no idea how this relates at all to my degree.
Ok thanks, that would be most appreciated. Thanks a lot for all the info I recieved so far from everybody.
I have made my eigenfaces but am actually confused about what to do with them. Initially I was just correlating them with the image that contained the faces I want to detect, however, this rarely works and does not seem like the right way to go. Basically I don't know what the next step is once the eigenfaces have been calculated.
That was what I was trying to do but it hardly works.
The eigen face here is from a training set of 4 images, although I did try a training set of 9 and the results were no better. I was using this eigen image and using the normxcorr2 function in matlab to correlate this image with a given image containing a number of faces. Very rarely did it work, although it was able to find the face on one image quite well.
Can you remove the shirt and tie? That looks quite a specific face, I think you need a bigger training set
Ok cool, I will use a bigger training set and give it another go tomorrow. However, my main confusion comes from what to do next once the eigenfaces have been obtained. I was simply correlating that eigenface with an image and looking for the highest peak which should correspond to a face. However this only worked in 2 images and so I am wondering if it was more luck than anything that this technique worked, as for the majority of cases it proved extrememly inaccurate.
I don't really understand what you mean?
Ok well that is far beyond my capabilities, I have no idea how to do that. I am a Mech Eng student, not an expert in computer programming.