Wormpose

_{Results videos adapted from Open Worm Movement Database license CC 4.0}

Overview

The WormPose package estimates the challenging poses of C. elegans in videos including coils and overlaps.

We train a convolutional neural network with synthetic worm images so that there is no need for human annotated labels.

Get started quickly

Try the tutorial notebook <a href="https://colab.research.google.com/github/iteal/wormpose/blob/main/examples/tutorial_sample_data.ipynb" target="_parent"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/></a>

This notebook goes over the whole WormPose pipeline with some sample data and an already trained model. You can run it in Google Colab.

Read the documentation

Check the Documentation website for detailed instructions.

Read the paper

WormPose: Image synthesis and convolutional networks for pose estimation in C. elegans
Hebert L, Ahamed T, Costa AC, O’Shaughnessy L, Stephens GJ (2021) WormPose: Image synthesis and convolutional networks for pose estimation in C. elegans. PLOS Computational Biology 17(4): e1008914. https://doi.org/10.1371/journal.pcbi.1008914

Manuscript data

Manuscript data is available here: https://wormpose.unit.oist.jp.