I'm super new to deep learning and computer vision, so this question may sound dumb.

In this link (https://github.com/GeorgeSeif/Semantic-Segmentation-Suite), there are pre-trained models (e.g., ResNet101) called front-end models. And they are used for feature extraction. I found these models are called backbone models/architectures generally. And the link says some of the main models (e.g. DeepLabV3 or PSPNet) rely on pre-trained ResNet.

Also, transfer learning is to take a model trained on a large dataset and transfer its knowledge to a smaller dataset, right?

  1. Do the main models that rely on pre-trained ResNet do transfer learning basically (like from ResNet to the main model)?

  2. If I use a pre-trained network, like ResNet101, as the backbone architecture of the main model(like U-Net or SegNet) for image segmentation, is it considered as transfer learning?


Your Answer

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

Browse other questions tagged or ask your own question.