deep_sort_pytorch

deep_sort_pytorch copied to clipboard

how to test the MOTA in MOT challenge

10

DENGZAIXU

前辈你好，请问要如何在视频流中既进行人的跟踪又进行识别呢，比如说识别人的戴帽子和戴口罩这种。在训练的时候已经加入了这些识别项

mozeqiu

zhaiwei-233

apply sort and not deepSort

1

Hi, thank you for this work! I want to run also sort and not deepsort. I think that I just need to make the match without features... Can you please...

olgapeled

i change detector ,use FCOS instead of YOlO, The effect of pedestrian detection is good, i want to Detect a frame to track 10 frames,it seem some trouble,can you show your code for real time processing

5

huangzicheng

Guys I have problems with installing the **requirements.txt** to setup the environment in order to be able to run deep_sort_pytorch. In particular I have encountered problems between opencv-python and numpy...

Lyubomyr01

间隔3，5，7，9帧，对deepsort的影响

6

间隔3，5，7，9帧，对deepsort的影响，有参数可以调节嘛？

starsky68

How to evaluate on MOT17 or MOT20 dataset?

8

The ground truth format has changed. Hence the same evaluation code won't work, right? I ran this implementation on both MOT15 and MOT17 datasets. The results are in the same...

Nishzzz

您好，请问你是怎么将这个修改成多类别的

3

.jpg?raw=true) 这样吗？这个需要修改 **yolov3_deepsort.py** ``` # select person class mask = cls_ids == 0 bbox_xywh = bbox_xywh[mask] # bbox dilation just in case bbox too small, delete this line if using...

susie-evil

HI! I convert the ckpt.pt mode to onnx. found the output feature different between torch and onnxruntime. If using "self.net.eval()", the result will be same. i confuse that dose the...

zhangcj13