Quantifying behavior is crucial for many applications. Videography provides easy methods for the observation and recording of animal behavior in diverse settings, yet extracting particular aspects of a behavior for further analysis can be highly time consuming.
Here we present an efficient method for 3D markerless pose estimation based on transfer learning with deep neural networks that achieves excellent results (i.e. you can match human labeling accuracy) with minimal training data (typically 50-200 frames). We demonstrate the versatility of this framework by tracking various body parts in multiple species across a broad collection of behaviors.
The package is open source, fast, robust, and can be used to compute 3D pose estimates. Please see the original paper and the latest work below.
Open Source Code:
On the analysis speed & robustness to video compression: bioRxiv(Oct 2018): On the inference speed and video-compression robustness of DeepLabCut
arXiv (April 2018): https://arxiv.org/abs/1804.03142v1 (Now published in Nature Neuroscience)
bioRxiv (Nov 2018): Using DeepLabCut for 3D markerless pose estimation across species and behaviors (now published in Nature Protocols)
For more information:
Alexander Mathis - firstname.lastname@example.org
Mackenzie Mathis - email@example.com
Info about Workshops and online video Tutorials!
Example use cases:
(click on the image to see more details and other use cases!)
*Launch the program “cmd” (Windows) or “terminal” (Mac) and in the folder where you downloaded the file type:
conda env create -f dlc-macOS-CPU.yaml
conda env create -f dlc-windowsCPU.yaml
conda env create -f dlc-windowsGPU.yaml
conda env create -f dlc-ubuntu-CPU.yaml
conda env create -f dlc-ubuntu-GPU.yaml
DeepLabCut in the News!