First poc

[almarklein]

Screenshot of the slicer_with_3_views.py exampe:

[emmanuelle]

Really cool! I'll read at least some of the code before our meeting this afternoon. Could you please add a minimal setup.py so that we can install the package locally to run the examples? (unless there is another preferred workflow)

[emmanuelle]

@nicolaskruchten 👀

[almarklein]

Could you please add a minimal setup.py

Will do. Though if you run python examples/xxxx.py from the root, it should also work, I think.

[almarklein]

Though if you run python examples/xxxx.py from the root, it should also work, I think.

Nope, the current dir is only in PYTHONPATH when running interactively. Anyway, setup.py is added, so you can do e.g. pip install -e .

[emmanuelle]

I know we sent you code where the image was added as a layout image but the more "modern" way of doing this is to use an Image trace like in https://github.com/plotly/dash-sample-apps/blob/master/apps/dash-covid-xray/app.py#L68 and https://github.com/plotly/dash-sample-apps/blob/master/apps/dash-covid-xray/app.py#L418. This way you can get hover and click events on image pixels, which is not the case with a layout image

[almarklein]

I saw that the two examples used different methods, and I assumed that layout images where easier, but will change this :) I think I'll do that in a new PR. Are there any other differences? E.g. I saw that with layout images one can create a stack of them and use alpha blending to overlay e.g. segmentation results. Is that possible with image traces as well?