I'm just started to learn about meta learning and CNN and in most paper that I've read they mention to have one CNN to feature extraction. These features will help the another network.

I don't know what is feature extraction (I don't know what are those features) but I'm wondering if I can use it on image segmentation.

The idea is to use the first network to feature extraction without doing image classification, and pass those features to the other network.

My question is: How can I use feature extraction in CNN on image segmentation?


1 Answer 1


Feature extraction is a way that people use pretrained model to extract information from input data. For example, image segmentation task may use the VGG network or other image classifying network for feature extraction. The output of the last convolution layer is taken. Then, the features are feed into the untrained network to get outputs. The bottom network for image segmentation usually consists of upsampling and convolutional layers. Then output of size of original image is resulted in teh main network. Hope I can help you


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .