Research
Our lab leverages deep learning and computer vision to study complex biological systems by building novel computational tools. Our core research areas of interest include:
Computational ethology
We develop methods for pose estimation and tracking to quantify animal behavior.
To help others make use of this technology, we created SLEAP: an open-source framework for the entire multi-animal pose tracking workflow.
Complex morphology phenotyping
We develop tools to quantify traits in complex morphologies such as plant root systems. We are applying these technologies to solve climate change in collaboration with experimental labs in the Salk Harnessing Plants Initiative.
Neuro-AI modeling
We are interested in exploring how artificial neural networks (ANNs) can be used as models of biological neural circuits. To do this, we combine agent-based modeling with imitation learning to take advantage of motion capture data to constrain our models. This work aims to improve our understanding of both brains and ANNs.
See Publications for a list of recent published works from the lab.