Training mtcnn in KITTI for vehicle detect

Open hkdqliu opened this issue 8 years ago • 3 comments

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?

Apr 19 '17 14:04 hkdqliu

@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.

Apr 20 '17 04:04 Seanlinx

@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

Apr 24 '17 10:04 hkdqliu

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

Jun 12 '18 02:06 ygl365166495