YOLOV4_MCMOT issues

Evaluating MOT Algorithm

Hi, I have an issue evaluating the tracker. Currently, I am getting the same track ids in a single frame for a different class. It seems that track IDs are...

sharoseali

Detection not working properly

Hello, I have trained yolov4-paspp from Pytorch Yolov4 Repo. I have tested it in that repo, and it works perfectly on my dataset. But when i test the weights and...

kenrickfernandes

@CaptainEven 打扰了，可以告知一下标签的格式细节呢，我明明使用的和mcmot一样的标签，但是，似乎一直出现越界的情况。比如我有54个类，标签是这样的： 0.0 1.0 0.18229166666666666 0.37962962962962965 0.028125 0.10648148148148148 0.0 2.0 0.6484375 0.3731481481481482 0.03333333333333333 0.09351851851851851 0.0 3.0 0.7234375 0.49722222222222223 0.028645833333333332 0.09444444444444444 5.0 1.0 0.18802083333333333 0.7231481481481481 0.28489583333333335 0.3527777777777778 5.0 2.0 0.521875...

starsky68

CUDA error: device-side assert triggered

1

Starting training for 100 epochs... Epoch gpu_mem GIoU obj cls reid total targets img_size 0%| | 0/1 [00:00

starsky68

Error in Multitracker.py while running demo.py

HI @CaptainEven I want to detect track 10 classes, for that I used my yolov3 detection model with .weights but when i run demo.py. I am getting the error below,...

sharoseali

Dear author, That's JDE implement method ?

I found this code same as jde algorithm, jde algorithm based anchor center points to generate reid feature, but when two objects corresponding same anchor, this prediction may be become...

Ronales

YOLOV4 Tiny model

@CaptainEven Can I get your pretrained model for YOLO? Or How can I use track COCO objects using this code?

YagnikThanki

Which branch should I use?

Thanks for putting this together, it looks like an excellent resource. However, I'm having some trouble getting the demo to work. All of the confidences are very low (

chrisrapson