[feature] support YOLOv5 and fix bugs

[cjliu01]

add some contents about YOLOv5 detector, tracking results showing and modify some errors about kalman_filter.py, detector.YOLOv3.weight.yolo_utils.py(This eliminates the need for compiling NMS operators).

[cjliu01]

later modifications to READMD.md will be uploaded, as well as updates to the appearance feature extraction network.

[ZQPei]

Thanks a lot for your contribution!
This PR title is a bit confusing. It will be much better to use a more descriptive title instead of ambiguous words like "some".

[cjliu01]

tracking

1. Added resnet network to the appearance feature extraction network in the deep folder
2. modified the NMS bug in the preprocessing.py and the updated covariance calculation bug in the kalmen_filter.py in the sort folder
   detecting
3. Added YOLOv5 detector, aligned interface, and added YOLOv5 related yaml configuration files
4. The training, evaluation and prediction in the original YOLOv5 were deleted
   deepsort
5. Added tracking target category, which can display both category and tracking ID simultaneously

[ZQPei]

tracking

1. Added resnet network to the appearance feature extraction network in the deep folder
2. modified the NMS bug in the preprocessing.py and the updated covariance calculation bug in the kalmen_filter.py in the sort folder
   detecting
3. Added YOLOv5 detector, aligned interface, and added YOLOv5 related yaml configuration files
4. The training, evaluation and prediction in the original YOLOv5 were deleted
   deepsort
5. Added tracking target category, which can display both category and tracking ID simultaneously

Great job! Will you update the title as well, please?

[ZQPei]

later modifications to READMD.md will be uploaded, as well as updates to the appearance feature extraction network.

I assume these commits will be uploaded soon, correct?
This pull request will be merged once the remaining is ready.

[cjliu01]

tracking

1. Added resnet network to the appearance feature extraction network in the deep folder
2. modified the NMS bug in the preprocessing.py and the updated covariance calculation bug in the kalmen_filter.py in the sort folder
   detecting
3. Added YOLOv5 detector, aligned interface, and added YOLOv5 related yaml configuration files
4. The training, evaluation and prediction in the original YOLOv5 were deleted
   deepsort
5. Added tracking target category, which can display both category and tracking ID simultaneously

Great job! Will you update the title as well, please?

Is there any standard for naming titles

[cjliu01]

later modifications to READMD.md will be uploaded, as well as updates to the appearance feature extraction network.

I assume these commits will be uploaded soon, correct? This pull request will be merged once the remaining is ready.

updated on the appearance feature extraction network and README.md

[ZQPei]

@cjliu01 Hi, it's me again.
I noticed that the previous PR(#262), that trying to add MaskRCNN to detector, now has conflicts with master.
Is it convenient for you to solve these conflicts that introduced by your PR with the previous one (#262)?
It will be really nice of you to help us out.

[cjliu01]

I will try to merge MaskRCNN into the main branch, but it will take some time.