I am a graduate student and I am working on image segmentation. I’m trying to create a simple image segmentation, but existing codes are quite difficult. Can you help me make a sample using existing data?
Do you have a specific dataset in mind?
Also, which examples/tutorials are you using and where are you stuck?
Here is a very simple example using just one conv layer for the model and some random input and output.
Fast.ai has introduced data block API which makes it really easy to create dataloaders and can then be used with Learner object. The course will be released as a MOOC in early 2019. Here is the notebook
Thank you very much for the answers. The data set I want to use is the ISPRS Vaihingen data set. I want to train the network using the Segnet model.