0
$\begingroup$

I have an idea to adapt YOLO algorithm to my application, the original YOLO algorithm is for image classifications, which have 24 convolutional layers with output class of 1000, is it possible to replace the basic network of YOLO with Alexnet or Resnet or the custom network structure designed by myself? Noted that my application have input shape of 500 * 10000 * 1 and only 4 classes for classification.

$\endgroup$
0
$\begingroup$

No, the original (or any) YOLO is for object detection. You can easily replace the feature extractor (DarkNet53, if I'm not mistaken) with any other, as long as you maintain the correct number of weights in the detection layer.

$\endgroup$
2
  • $\begingroup$ The basic network behind YOLO is CNN, why it's cannot be replaced? $\endgroup$ Jan 15 at 8:29
  • $\begingroup$ Pls read my answer again $\endgroup$
    – Alex
    Jan 15 at 11:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.