Hello,
I'm having a hard time finding an example of how to implement a convolutional neural network for image semantic segmentation in R. An example of such a network is a U-Net developed by Olaf Ronneberger, Philipp Fischer and Thomas Brox. Specifically I'm having difficulties understanding how I can load batches of images and corresponding masks into the neural network.
In my case the images are stored as .jpg files in a directory and the masks as data frames in R.
One image can have multiple masks, but I would be already happy if I can train a neural network with one mask per image.
A good example would have been this: unet_example
Unfortunately, running this example on my computer yields an error for which I have created an issue here: keras_team_issue.
Can anyone point me to some good beginner resources, preferably R-code?
Please, don't make me have to turn to Python...
Many thanks!
Denis
edit: typos