the speed of this paper, i have some questions. Please help me.

[oujieww]

in your paper's abstract ,"the fastest real time system with ∼23 frames/sec". And we can find "Keypoint and person detections take 35 ms while PRN takes 2 ms per instance" in 4.5 runtime analysis ,So we can get a result 1000/(40) ~=23 fps.

my questions are : 1."is the 35ms include "load image, resize image, transform image to tensor, normlized the image , put the image to cuda, and model inference" ? or "35 ms only for model inference?"

2.i test the speed of PRN , is same as reported on paper 2ms/instance ,but how about the time for "select the box from the retinanet" , "crop the feature map for every instance " and "resize the every instance "

i try to implementation whole repo by pytorch, but "load image , resize , to tensor ....." "select box , nms, crop heatmaps for everyone " cost too many time..., So ,how long your code take for those operation? ^-^,Thanks...!

[oujieww]

and how about the time for "cuda heatmaps for instance transformed to keypoints coordinate"