Omnipose is a general image segmentation tool that builds on Cellpose in a number of ways described in our paper. It works for both 2D and 3D images and on any imaging modality or cell shape, so long as you train it on representative images. We have several pre-trained models for:
bacterial phase contrast: trained on a diverse range of bacterial species and morphologies.
bacterial fluorescence: trained on the subset of the phase data that had a membrane or cytosol tag.
C. elegans: trained on a couple OpenWorm videos and the BBBC010 alive/dead assay. We are working on expanding this significantly with the help of other labs contributing ground-truth data.
cyto2: trained on user data submitted through the Cellpose GUI. Very diverse data, but not necessarily the best quality. This model can be a good starting point for users making their own ground-truth datasets.
Here we provide both the documentation for Omnipose and our fork of Cellpose. Please note this documentation is actively in development. For support, submit an issue on the Omnipose repo. For more on the workings of cellpose, check out our twitter thread and read the paper.