Jimmy Chen

I am double checking the problem as I have a very large main memory (64G). May I ask the typical machine configuration (CPU, GPU) on which the code runs well?...

Hi, thanks for interested in our work. I may first write a python version for the matlab code in: https://github.com/lood339/two_point_calib/tree/master/matlab The python version is here: https://github.com/lood339/two_point_calib/tree/master/python

For the parameters of the decision trees, I add two files in https://github.com/lood339/two_point_calib/tree/master/data Thank you.

For testing, you have to train the random forest first. printf("modelFile: a random forest . \n"); This is random forest model. printf("ptz_feature_folder: .mat file contains ptz, keypoint and descriptor.\n"); This...

For the projection, it should be straightforward. Given pan-tilt-zoom parameters, we can get the projection matrix. Then, we can project the pitch to the image. A C++ reference code is...

Thanks for interested in our work. For the Rugby dataset, you have to prepare following data. 1. Court model. It is the geometry information of the Rugby field. For example,...

Sure, but it is not clear to me which .mat file you refer to. Do you mean some .mat files in https://github.com/lood339/SCCvSD/tree/master/data? There is a readme file for the data...

Sure, you can have a look here: https://github.com/lood339/SCCvSD/issues/4 The idea is that homogrpahy--> image 2D- world 2D correspondence --> estimate all parameters of the camera (location, rotation and focal length)....

Can I have more information of "disconnected lines output"? I assume you use the soccer dataset. One way is to check if the training is converged. For example, how about...

1. The edge detection in the training set does not look good. I suspect the training process does not converge. For the training set, the network should give very good...