0
$\begingroup$

I want to develop face recognition program using convolutional neural network. Can some one tell me steps to follow to do the same? I am new to deep learning.

I want to develop it on windows using python and we can use Keras or Tensor Flow . I want to do live detection.Like first i will store the photos of me and another person in a folder .I will train and finally when i want to run the script so that people on webcam are classified with 95% accuracy .I did this using ML algorithm and got 80% accuracy but i want 95% accuracy .I have done Andrew Ng's course on AI.Note we can use Ubuntu also as i have both OS on my machine .My machine does not have a GPU . Also procedure should not be simplest and accurate possible but close to that .

Problem i am facing - I don't know how to write the code for neural network .I just know the concept .

$\endgroup$

closed as too broad by respectful, nbro, John Doucette, Oliver Mason, DukeZhou Oct 10 at 20:43

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 4
    $\begingroup$ It's not really possible to give a step-by-step guide as there are several ways to do face recognition, and which kind of system you want to develop will depend a lot on how it is going to be used. Instead I suggest you take a course to give you a better background in the subject. The course coursera.org/learn/convolutional-neural-networks might be a good place to look, as it includes some material on face recognition including a workshop where you make some classifiers. If you are new to deep learning then you may want to take some of the earlier courses too. $\endgroup$ – Neil Slater Oct 8 at 19:41
  • $\begingroup$ @NeilSlater If there are many ways then you can tell simplest and most accurate one .Knowing something is better than knowing nothing . $\endgroup$ – Brandon Oct 9 at 7:24
  • $\begingroup$ There is no "simplest and most accurate" system in general for face recognition. If you could narrow down your scenario and give lots of specifics, including nature of images that will be tested, what training data you have access to, what training system(s) you have access to, what your production system is (i.e. on what devices your algorithm needs to run), then perhaps someone could point you in a very general way to a type of algorithm and library to use. But not a step-by-step recipe. Otherwise I suggest you take a course, because then you can come back with some precise questions. $\endgroup$ – Neil Slater Oct 9 at 8:02
  • $\begingroup$ @NeilSlater I want to develop it on windows using python and we can use Keras or Tensor Flow . I want to do live detection.Like first i will store the photos of me and another person in a folder .I will train and finally when i want to run the script so that people on webcam are classified with 95% accuracy .I did this using ML algorithm and got 80% accuracy but i want 95% accuracy .I have done Andrew Ng's course on AI.Note we can use Ubuntu also as i have both OS on my machine .My machine does not have a GPU . Also procedure should not be simplest and accurate possible but close to that . $\endgroup$ – Brandon Oct 9 at 8:09
  • $\begingroup$ I suggest you use edit to add those details to the question before it gets closed for being too broad. Also, if you have done Andrew Ng's course, please explain what is preventing you simply using the face recognition approach described there, which seems very close to what you want to do, including using Python, TensorFlow. I would assume you are using triplet loss and calculating embeddings for your identities? Getting a low accuracy may be entirely different problem, and does not mean you need to suddenly cast your net wide and ask about all possible face recognitoon algorithms. $\endgroup$ – Neil Slater Oct 9 at 8:18