The dnn module has been added into digiKam. I have finished the code about algorithm selection, database modification and recognition classes. I met a lot of compilation errors which is caused by the incompatible between the new code and digikam. But the errors are solved with my mentors' help.
Now the accuracy in face recognition in digikam has reached 99%. Here is the screenshot of the face recognition process.
Before recognition:

Images in tag "jay" after recognition:

Images in tag "Robert" after recognition:

Images in tag "unknown" after recognition:

Plan for next step:
(1) Speed up the dnn module, it will be better that it can be as fast as dlib.
(2) Polish and document the new code in dnn module.
(3) Finish face region database synchronization which is proposed in my proposal.