Following the theme with RunCellpose, this plugin provides a wrapper around StarDist's deep learning segmentation models. Current features:
Supports loading of inbuilt StarDist models.
Supports loading of custom models.
Will run on a GPU if configured.
Works in 2D or 3D. The latter requires you to train your own model.
Can run on brightfield colour images.
User can capture the probability images.
Provided settings to tile input images, allowing large images to be processed without running out of memory.
Current interface:
Current figure display (with probability export on):
The dependencies can be a bit of a pain to install, but I think the instructions should be clear enough.
An outstanding issue is that the GPU integration seems to struggle with multiple worker processes. Unlike Pytorch, running out of GPU memory doesn't trigger a hard error and so stalls the worker process. Not sure how to fix that.
Following the theme with RunCellpose, this plugin provides a wrapper around StarDist's deep learning segmentation models. Current features:
Current interface:
Current figure display (with probability export on):
The dependencies can be a bit of a pain to install, but I think the instructions should be clear enough.
An outstanding issue is that the GPU integration seems to struggle with multiple worker processes. Unlike Pytorch, running out of GPU memory doesn't trigger a hard error and so stalls the worker process. Not sure how to fix that.