search

shaileshmulange / javacv

Automatically exported from code.google.com/p/javacv
GNU General Public License v2.0
0 stars 0 forks source link

Face Recognition using Eigen and Fisher Face. #300

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
I am trying to do face recognition using Eigen and Fisher Face Method..
But I got this error:
" OpenCV Error: Unsupported format or combination of formats (In the 
Fisherfaces method all input samples (training images) must be of equal size! 
Expected 10404 pixels, but was 2025 pixels.)"

Same Recognition codes works fine on LBPH but the results are not good..

So need help with this case..

Original issue reported on code.google.com by hnfarid...@gmail.com on 9 Apr 2013 at 11:11

GoogleCodeExporter commented 9 years ago 
This probably has the same cause as issue #294, i.e.: Try to keep a Java 
reference to all your images.

Original comment by samuel.a...@gmail.com on 11 Apr 2013 at 10:44

GoogleCodeExporter commented 9 years ago 
I am doing the same project,
btw, how you do face recognition using Eigen and Fisher Face Method?
I can't do the fisher part

I'll appreciate your respond

Original comment by muhammad...@gmail.com on 22 Jun 2013 at 7:46

GoogleCodeExporter commented 9 years ago 
I'm not sure why I mentioned issue #294, it doesn't sound related. Anyway, we 
can do face recognition as shown here:
http://pcbje.com/2012/12/doing-face-recognition-with-javacv/
But it's a bit buggy. We need to keep a Java reference to all IplImage objects 
created, or they will get garbage collected, and memory errors will occur.

Does it run properly, or do you still have issues with that?

Original comment by samuel.a...@gmail.com on 16 Sep 2013 at 11:21

GoogleCodeExporter commented 9 years ago 
as memory issue mentioned by Samuel, yes I am continuously getting memory or 
null reference issue if trying to use the example . Is there any permanent 
solution available..
getting this error-
http://answers.opencv.org/question/22163/android-error-fatal-signal-11-sigsegv-a
t/

and all available alternatives mentioned failed in Nexus 4

Original comment by abhishek...@gmail.com on 11 Oct 2013 at 8:15

GoogleCodeExporter commented 9 years ago 
Thanks for all who responded, Specially Samuel :)
U helped a lot..
Because of your guidance I did my project..
Thanks :)
Regards,
Hammad

Original comment by hnfarid...@gmail.com on 12 Oct 2013 at 7:01

GoogleCodeExporter commented 9 years ago 
@abhishek Simply keep a live reference to `grayImg`, for example in an array or 
something, and that should fix the issue.

So, it looks like that works around this issue, but I agree this needs some 
work...

Original comment by samuel.a...@gmail.com on 4 Jan 2014 at 2:12

GoogleCodeExporter commented 9 years ago 
Issue 416 has been merged into this issue.

Original comment by samuel.a...@gmail.com on 31 Jan 2014 at 2:32

GoogleCodeExporter commented 9 years ago 
Since JavaCPP/JavaCV 0.8, the mapping of the C++ API uses `cv::Mat` instead of 
`IplImage`, so this memory management issue no longer exists. Enjoy!

Original comment by samuel.a...@gmail.com on 30 Jul 2014 at 11:31