This is the repository of my research into the use of monocular pose-estimation technologies in estimating and tracking the pose of a cheetah. The repository is forked from Open MMLab-MMPose.
The Jupyter Notebook used in developing object detection models is similarly forked from Open MMLab-MMDetection, and can be found here.
The Jupyter Notebook file, AnimalPose.ipynb, contains all project-specific code. Function definitions (including MMPose function adaptations) can be found in setup_env.py.
Config files for 2D pose-estimation are located here.
Config files for temporal pose-lifting are located here.