Towards-Realtime-MOT

Towards-Realtime-MOT copied to clipboard

ZeroDivisionError: division by zero

3

When I finish training my data set, I get **latest.pt**, errors are reported when tracking with this weight as follows: File "E:\python_workspace\JDE\track.py", line 203, in main(opt, File "E:\python_workspace\JDE\track.py", line 139,...

jzp-skam

Thanks for your awesome research! I have a quick question about this project. can I use the model with real-stream like camera?

Wooho-Moon

can train multi-class Tracking ?

1

能否进行多类别的跟踪？

xuaixi

How to run with camera.

5

Can anyone show me how to run with camera or webcam? Many thanks.

CommandoHuan

Processed cuhk-sysu has 11.7k id's but original cuhk-sysu has 8.4k id's ?

4

sachingoyal0104

What's the correct value of "conf" during training?

2

Hello, thanks for your excellent work When I training in VisDrone dataset, the conf during training will decrease to around 0.002 after 35 Epochs, and the Precision is good but...

JackWoo0831

作者您好，很有幸能拜读您的代码。但是我有一个小问题想向您请教，“self.lost_stracks = sub_stracks(self.lost_stracks, self.removed_stracks)”， 这是在您的代码mutitracker中的第336行。这里您采用将两个轨迹集合相减的方法以去除self.lost_stracks轨迹集合中的removed的轨迹，但是在此时 self.removed_stracks中并没有对当前帧的removed的轨迹进行添加，这样会不会造成self.lost_stracks轨迹集合中无法去除当前帧新生成的removed轨迹？

Xavier-Lin

Hello @Zhongdao Why the embedding representation of appearance is D*H*W, how can so many representations distinguish the representation of positive and negative samples？

APeiZou

Hi. Thank you for your great work. I just want to ask if you did minus the center_x and center_y by 1. Because the MOT dataset's coordinates are 1-based indexing,...

sieumap43

mAP is always zero

34

Hi, when I try to train the network with the provided data, the mAP is always zero! Is it normal? How many iterations does it take to reach the higher...

Marjanmoodi