I have a question in regards to merging features extracted from CNN and handcrafted features. I have been reading this paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9002895/#B33-sensors-22-02467 I am not sure if I am missing something but it seems that features are extracted from a CNN (ResNet34 extracts 512 features) which are passed from a sigmoid and then merged with handcrafted features (11 features). Then the merged features are passed from a fc layer (1000) and then perform binary classification based on the output.

My question is, what is the 1000 in the fc layer represent because I would expected that we have 523 features passed in the fc layer and the output to be binary.

Thank you



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.