Deep sort with yolov7 and yolov5 shape error

[SalmaZakaria]

I have tried to use both yolov7 and yolov5 with deep sort and every time it gives me this error In yolov7: ValueError: shapes (2,352,3) and (3,352,3) not aligned: 3 (dim 2) != 352 (dim 1) In yolov5: ValueError: shapes (2,640,3) and (3,640,1) not aligned: 3 (dim 2) != 640 (dim 1)

[JKMCCALLA]

Feature vector has to be one dimensions along one axis (a list of numbers)

For example: 128X1 (as mentioned in the research paper)

This article should help break down the steps for image characterization: https://pyimagesearch.com/2014/03/03/charizard-explains-describe-quantify-image-using-feature-vectors/#:~:text=Image%20Feature%20Vector%3A%20An%20abstraction,used%20to%20represent%20an%20image.

@SalmaZakaria

[pnthai88]

Feature vector has to be one dimensions along one axis (a list of numbers)

For example: 128X1 (as mentioned in the research paper)

This article should help break down the steps for image characterization: https://pyimagesearch.com/2014/03/03/charizard-explains-describe-quantify-image-using-feature-vectors/#:~:text=Image%20Feature%20Vector%3A%20An%20abstraction,used%20to%20represent%20an%20image.

@SalmaZakaria

Thanks for instruction. The correct of my problem is convert frame to gray

[JKMCCALLA]

Welcome. Great!