As of arivis Pro 4.1.1 (formerly Vision4D), Cellpose-like segmentation is supported natively within the analysis pipeline. However, arivis Pro does not use the native Cellpose model format and instead relies on the ONNX standard. Therefore, Cellpose pretrained and custom models must be converted to the ONNX format before they can be used in arivis Pro.

We have been providing a Python script for this conversion for some time now. To make it easier for you to use this script and to also support Cellpose 3 and Cellpose 4 models, we are now offering the Cellpose to ONNX Converter tool that allows for easy operation via a local web interface.

Prerequisite: To run the tool, Docker Desktop must be installed and running on your computer.

Please note: Cellpose models may be subject to usage restrictions, including limitations on commercial use. Before downloading or using a model, review the applicable license terms carefully. For more information, see the https://github.com/mouseland/cellpose