Closed td042 closed 6 years ago
Could you please describe the details about the architecture you used for the dnn mmod face detector? e.g. the depth of the network or the activation function. Which dataset did you use for training?
All this is documented here: http://dlib.net/dnn_mmod_face_detection_ex.cpp.html
IF YOU ARE REPORTING A BUG OR PROBLEM WITH DLIB THEN FILL OUT THE ENTIRE TEMPLATE BELOW. ISSUES ASKING QUESTIONS ABOUT WHY SOMETHING DOESN'T WORK THAT FAIL TO FILL OUT THE ENTIRE TEMPLATE WILL BE CLOSED.
It is OK to suggest interesting improvements to dlib, even if you are not volunteering to implement them. However, the issue tracker is not a code writing service, do not ask for someone to write code for you. E.g. Do not ask for feature improvements to the example programs. If there is some feature improvement you want in an example program then it's up to you to write it.
Before you ask a question, check Google for a solution, the dlib FAQ, or consult the dlib documentation. Every single function in dlib is documented in detail. If you obviously haven't read the documentation your issue will be closed.
If you aren't reporting a bug or problem with dlib then delete this template and write whatever you want here.
Expected Behavior
Current Behavior
Steps to Reproduce