TengfeiZhang

I want to know which way do you train the model? Tf or caffe? And If my processing is inconsistent with the pre-trained model, will it affect performance?

@chiayuhsu Did you solve this problem？

@michelleweii non_max_suppression_fast()这个函数的参数没对上！你改下就好了！ non_max_suppression_fast(np.hstack((box, np.array([[p] for p in np.array(probs[key])]))), overlap_thresh=0.5)

@michelleweii 把你的bboxProb组织到一起 [x1,x2,y1,y2, prob]