Handle instance segmentation by adding second output channel for touching borders

Nobody has claimed this yet.

At the moment we don't do instance segmentation. This allows our model to be very simple while still achieving amazing results for our current use-cases. Where it breaks down, though, is use cases like extracting buildings in very densely populated areas.

See for example a quick prototype for Tanzania where the segmentation mask alone can not distinguish between touching buildings.


And while proper instance segmentation models are much more complicated, there is one trick we can pull off based on what the folks in https://arxiv.org/abs/1806.00844 propose.
- [ ] Add a second channel to the output. The first channel will be the segmentation mask as it is right now. The second channel will represent touching features - and only the border between features.
- [ ] Train with ground truth segmentation masks and compute and rasterize borders where features touch.
- [ ] After prediction feed results through the watershed transform to divide touching features into multiple features.

As a result we will get instance segmentation and can distinguish between buildings in the use-case above.

Start by reading the linked paper and the issue checklist, focusing on the proposed second output channel and watershed transform. No repository files or tests are named, so the implementation entry points must be located first. Done means training can produce touching-border masks and post-processing separates touching buildings into distinct instances.

Written by the indexing model from the issue text.