Training using Matlab face detector or voila jones algorithm.

[nishantkapur]

Greeting....I have been trying to train the model using matlab's face detector.....but i have not been able to achieve anything. I see that in you loadsamples.m you have given an option for using cascade classifier but some how i cant understand how to use it. Also I coudnt understand the difference between data{}.bbox_gt and data{}.bbox_facedet.

Thank you in advance .

[jwyang]

Hi, you can use matlab detector to obtain the face region, and then assign it to data{}.bbox_facedet. In my current version of code, I commited that part for clarification. For your needs, you can remove the comments from line 53 to 74. And change line 95 to:

if ~isempty(ind) | ~isdetected(i)

Then you can train the model you want!

best,

At 2015-01-28 18:09:04, "nsihantkapur" [email protected] wrote:

Greeting....I have been trying to train the model using matlab's face detector.....but i have not been able to achieve anything. I see that in you loadsamples.m you have given an option for using cascade classifier but some how i cant understand how to use it. Also I coudnt understand the difference between data{}.bbox_gt and data{}.bbox_facedet.

Thank you in advance .

¡ª Reply to this email directly or view it on GitHub.

[nishantkapur]

Hey, thank you for your help. I am trying to implement this paper in C++. i have some doubts regarding image transformations. You have used non reflective similarity(fitgeotrans/cp2tform) and fwdtform function. My doubt is is non eflective similarity equalent to affine transform in opencv? if not then what is?. also whats the alternative to fwdtform of matlab function in opencv C++. Thank you.