Mohamed Lamnouar

Thank you for reply, ok I have two question : 1)- We guess our result are correct how can I modify it for continue the steps related to optimized NCNN,...

thank you for reply just I have followed all step which describe in section MegEngine 1)- optimization of YoloX using MegEngine On CPU.  2)- Yolox without optimized on CPU...

Hi, Did you solved this problem ?

Hi sorry for deleting my previous answer because my previous answer is for another question like yours, I think problem come from when you resize image. (when I was working...

yes, I think no option in this repository but you need to code a little bit. (take care about range of you number in you file because inside byteTrack there...

if I understand your question the FPS just for part tracking not inclue detection (i.e not inclue yolox)

Hi, yeah, I agree with you, I think just we need to add some modification like increment number of lost of this person of course if Stracks not empty, and...

I think this problem due to person re-identification not perfect you have two option if you need to solve this issue : 1. change the model of person re-identificatioon (recommended)...

Hi, you can press on button "q" in keyboard.

@AhmedKhaled945 I think the only one for solving this issue using Re-id, but you will another problem is you have some vehicle are same, but you can use the topology...