I reproduced the original version of yolov3 / yolov4, but encountered the following problems:
1、The automatic differential derivation architecture is not fast (such as GIOU);
2、The network layer is not easy to control freely, and the location of the retrieval layer Very inefficient;
3、Network inference is much slower than the original framework (even with GPU).
yolov3-yolov4-matlab respository here: https://github.com/cuixing158/yolov3-yolov4-matlab
I hope mathworks can strengthen the flexibility and efficiency of the above aspects ,thankyou!
I reproduced the original version of yolov3 / yolov4, but encountered the following problems: 1、The automatic differential derivation architecture is not fast (such as GIOU); 2、The network layer is not easy to control freely, and the location of the retrieval layer Very inefficient; 3、Network inference is much slower than the original framework (even with GPU). yolov3-yolov4-matlab respository here: https://github.com/cuixing158/yolov3-yolov4-matlab
I hope mathworks can strengthen the flexibility and efficiency of the above aspects ,thankyou!