neural network

[hodaghezavati]

hi, i want to ask what kind of neural network you used in order to detect human or face in your project and how did you train it?

[prsntmaurya]

Hello, we haven't used any neural networks. For the purpose of face detection, HAAR features are used and HOG features are used for Person detection. Please refer to the shared project report for more information.

[hodaghezavati]

hi, thank you for answering me i have found your project quite great may i ask one more question please? for labeling detected faces can i give a name? for example i give my name to my picture in dataset and then do the training. is it possible? where can i make the changes?

…

On Fri, Jan 11, 2019 at 10:43 AM prashant maurya ***@***.***> wrote: Hello, we haven't used any neural networks. For the purpose of face detection, HAAR features are used and HOG features are used for Person detection. Please refer to the shared project report for more information. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#41 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AgzrZ1NIeSo39FNinh-5kq4y6gfgTGjyks5vCDmzgaJpZM4Z1wgO> .

[prsntmaurya]

You can use custom names for images but for that, you have to make certain changes in the code create_face_model.py.
The current create_face_model.py accepts the images in the following format:
subject01.1.jpg
subject01.2.jpg
...
subject01.100.jpg

subject02.1.jpg
...
subject02.100.jpg

Here the number just after the word "subject" is class ie: it indicates the person to which image belongs to.
Example:
subject01, subject02 are two different persons.
subject01.1.jpg, subject01.2.jpg are images of same persons ie: subject01
subject01.1.jpg, subject02.1.jpg are images of different persons ie: subject01

[hodaghezavati]

Thank you very much for your explanation.

…

On Fri, Jan 11, 2019 at 1:49 PM prashant maurya ***@***.***> wrote: You can use custom names for images but for that, you have to make certain changes in the code *create_face_model.py*. The current *create_face_model.py* accepts the images in the following format: subject01.1.jpg subject01.2.jpg ... subject01.100.jpg subject01.1.jpg — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#41 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AgzrZ7lMDZ-xMl44aqsMQ8gUaJU8nzJeks5vCGUYgaJpZM4Z1wgO> .