Gil Levi
In which line does the error occur?
Hi, Thank you for your interest in our work. The original word was done in Caffe (prototxt files work with Caffe). For questions on the TensorFlow implementation, please refer to...
Hi, I'm afraid we do not have the original, uncropped images.
Hi, I am not sure what you are referring to, there is no use of fiducial detector in my code. Best, Gil
I see, try to open an issue in that github, I didn't use that code.
Hi, Thank you for your interest in our work. The common pipeline for using the network is applying face detection on the image and feeding the detected network (ideally, after...
In the above case, I suggest running a (human) face detector and run the network on the faces it detected in the image. Since it won't probably detect the dog...
Not in the near future.
Thank you for your comment, I ignored those cases in my code and training (see lines 80 and 166 in create_train_val_txt_files.py).
Hi Adel, Thank you for your interest in our work. If you are getting 30% error then you are actually getting 70% success which is not bad at all (since...