DIPLOMAT
Deep Learning-Based Identity Preserving Labeled-Object Multi-Animal Tracking
DIPLOMAT
Deep Learning-Based Identity Preserving Labeled-Object Multi-Animal Tracking
DIPLOMAT
DIPLOMAT provides a multi-animal pose estimation and editing interface. It relies on a trained CNN model (currently supporting SLEAP and DeepLabCut packages) and uses algorithms to first Track the animal body part in a way that reduces body part losses and identity swaps, and then provides an intuitive and memory/time efficient Interact interface to edit and re-track as needed. DIPLOMAT differs from other multi-animal tracking packages by working directly off of confidence maps instead of running peak detection, allowing for more nuanced tracking results.
Here we can make precise adjustments to our model's predicted points and re-run the Viterbi algorithm
Learn more about DIPLOMAT's functionalities through our video guides.
Introduction on how you can track and interact with pose data in DIPLOMAT, leveraging models created through SLEAP or DeepLabCut (DLC).
Overview of the visual settings, continuous labeling, single/multi body part labeling modes, and re-running the Viterbi algorithm
The Tweak Interface allows for fine-tuning to minimize identity swaps and other issues. Explore the tutorial video to learn more.
Tutorial on navigating the Tweak Interface