I want to test the face recognition code from the following page
https://docs.opencv.org/2.4/modules/contrib/doc/facerec/tutorial/facerec_video_recognition.html
Compiling was successful, so I made face image dataset which consists of my pictures and several pics of random people on the internet. But the accuracy was very poor.
Can you please tell me how to improve the accuracy of face recognition? I don't have any ideas.
Hi Kristi,
I think, you should use more pictures in your dataset. Usually, it leads to better results.
And you can also try to use some image processing to your dataset that you must have created, like histogram equalization on openCV...?
Thanks a lot. I followed your advice and enlarged the number of pictures in the dataset. It helped a little. Are there any other tricks that can be used to improve results besides histogram equalization?
Just drop us an email to ...
Show more 
Log in with Google