search

zhaoweicai / mscnn

Caffe implementation of our multi-scale object detection framework
405 stars 211 forks source link

Performance look wired in "mscnn-7s-720" compared to "ms-7s-720-pretrained" in Caltech #27

Closed worm004 closed 8 years ago

worm004 commented 8 years ago

hi @zhaoweicai,

I tested "ms-7s-720-pretrained" in Caltech-usa-test in reasonable and got great performance 10% (log-average miss rate) as the record in the paper.

Then I tried to train a new model based on "mscnn-7s-720". The training data is Caltech-train04 which includes 32077 images. After training, I got "mscnn_caltech_train_2nd_iter_25000.caffemodel" in the "mscnn-72-720" directory. However the performance is 0.89 (log-average miss rate). The reason may be the training data but I am not sure. Could you help me, thanks.

worm004 commented 8 years ago

Sorry, I forgot to set do_bb_norm to 1 when generate detect rects. After doing that I got 11.1%, pretty close to 10.0%. Is there anything else I can do to get the performance of the pre-trained model? Thanks.