Something i don't understand?

[lunalulu]

thanks for your wonderful jobs~ I have some questions that I don't understand. I hope you can help me answer them. 1、src/demo.py line26-31, I don't know why I'm doing this with the input? 2、src/demo.py line72，I want to know why do I have to add 4？

thanks for your reply!

[lunalulu]

？

[wangziren1]

？

Sorry, that is misoperation. Next, I will answer your question. 1.Input videos has different resolution, so they need to be processed to have same size as network input. ie. 384*384. You can modify it according to your input. 2.end_y[i] - s_y and end_x[i] - s_x are relative location, and are centered on s_y and s_x. They need to be transformed to take top left corner of 9 * 9 grids as start point.

[lunalulu]

@wangziren1 thanks for reply 1、The purpose of np.concatenate is to ensure that the scale of the human being in the image does not change during scaling, and changing the aspect ratio to 1:1?? 2、i think End_y [I] -s_y and end_x[I] -s_x are both coordinates relative to 12*12, as long as the difference between them is all within 9.