wuhao5688

In the Yolo paper, Yolo actually uses 7x7 cells for predicting bboxes and doing classification. So after performing several convs and poolings, the activation maps will have the dimension of...

I guess you have run the 'demo.py' to generate the prediction. In the function 'process_predicts()', it only select the cell having the highest confidence, i.e., P(object) * IoU * P(class|object),...

Sure, I adopted the map matching algorithm "Hidden markov map matching through noise and sparse- ness". I'll update the map matching code later. ( I'm occupied with a new deadline...

Hi, I'm sorry for letting you waiting so long. Now the codes are done. Note that my code base is a little bit large which makes me hard to clean...