Bug in face_preprocess_10kUS/create_negative_24c.py? about cnn_face_detection

[anson0910]

The train_val.prototxt files of the calibration nets in CNN_face_detection_models were changed to have a false mirror property in the data layer.

[anson0910]

Yes, I think that is a mistake!
Sorry... my calibration models were trained improperly, and I currently haven't found time to train them correctly, but I have fixed the training .prototxt files.

Previous ROC results on the FDDB were about 10% lower than the original work, but I think fixing the calibration models will reduce the difference.

[hiyijian]

sorry, I didnt see any change in training .prototxt.

[hiyijian]

yes. I noticed that issuse you guys discussed previously.
One more question please: line 560-563 in face_detection/face_detection_functions.py, both current_x and current_y time 2, why is that? As far as I am concerned, current_x and current_y are exactly the coord of top-left corner of the proposed window. there is no need to times 2?

[anson0910]

The 2 corresponds to the pooling layer of the first 12 net, since the pooling layer scales the input image down by a factor of 2.

[hiyijian]

thank you