Parry26

Tuning params to improve detection and reduce false positives

2

I have trained a fast r-CNN on my own dataset and it is working quite well (about 80% accurate for the most probable bbox). I now want to tune it...

Thanks for your great implementation. Do you have any thoughts on why DD in Keras produces such different results from the Caffe version? Specifically, in the Caffe version, whole objects...