2
$\begingroup$

What is the difference between using CNN with handcrafted features and CNN without handcrafted features?

Thank you

$\endgroup$
2
$\begingroup$

A CNN automatically extracts features, so hand-crafting features has become unnecessary for most applications. It learns what features to extract via backpropagation.

You can introduce handcrafted features in a few ways to boost the performance of a CNN:

  • inject hand-crafted features into the fully connected layer
  • feed the CNN output into another classifier alongside the hand-crafted features

This paper explores such concepts.

| improve this answer | |
$\endgroup$

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.