search

Seanlinx / mtcnn

593 stars 264 forks source link

Training mtcnn in KITTI for vehicle detect #34

Open hkdqliu opened 7 years ago

hkdqliu commented 7 years ago

Hi, @Seanlinx I have trained mtcnn in KITTI for vehicle detect, I only use the samples contain some cars(only for car detect). I founed that the detction result is so bad. When I use the trained P_net to produce the R_net‘s trained samples, the pos: part: neg = 0.8 : 10: 20, the positive samples's percentage is not balanced, what's your advice to change it?

Seanlinx commented 7 years ago

@hkdqliu hi, the ratio of pos:part:neg I used in training mtcnn Rnet is about 1:1:2. But I don't think the network structure of mtcnn is suitable for car detection for the aspect ratio of car bboxes varies.

hkdqliu commented 7 years ago

@Seanlinx I changed the structure of mtcnn to 1:2(height : width). the positive samples produced by Pnet is much less than neg/part ones

ygl365166495 commented 6 years ago

Maybe you need remove 'convert_to_square' function and change the 'cellsize' of the 'core/MtcnnDetector.py'.