We are excited to share our latest publication, developed in the context of our RTG 3144.
The work presents a deep learning-based approach for robust 3D tracking and individual identification of zebrafish using a two-camera setup. Our results show that robust tracking of up to 10 zebrafish over a period of 30 minutes is feasible, while maintaining individual identities over time. This provides a basis for long-term, quantitative analysis of individual locomotion and behavioral patterns—an important step toward more automated and objective behavioral experiments.
We congratulate all authors!