Metadata-Version: 2.1
Name: wf-pose-tracking-3d
Version: 0.1.0
Summary: Classes and methods for performing 3D pose tracking from 2D poses
Home-page: https://github.com/WildflowerSchools/wf-pose-tracking-3d
Author: Theodore Quinn
Author-email: ted.quinn@wildflowerschools.org
License: UNKNOWN
Keywords: cv
Platform: UNKNOWN
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python
Requires-Dist: wf-cv-utils (>=0.1.0)
Requires-Dist: wf-cv-datetime-utils (>=0.1.0)
Requires-Dist: wf-smc-kalman (>=0.1.0)
Requires-Dist: numpy (>=1.14)
Requires-Dist: scipy (>=1.1)
Requires-Dist: pandas (>=0.23)
Requires-Dist: matplotlib (>=2.2)
Requires-Dist: networkx (>=2.1)
Requires-Dist: boto3 (>=1.7)

# wf-pose-tracking-3d

Classes and methods for performing 3D pose tracking from 2D poses:

* Functions which load 2D pose data from Wildflower S3 directories

* Functions which perform very basic 3D pose reconstruction from multi-camera 2D pose data

* Functions which support basic visualization of 2D and 3D pose data

## Testing

To run tests, change working directory to `/tests` and run one of the following.

`python 3d_reconstruction_example.py` pulls a single frame of 2D pose data from S3 and performs a simple 3D pose reconstruction

`python 3d_pose_tracking_example.py` pulls multiple frames of 2D pose data from a local file and builds a set of 3D pose tracks

These scripts pull camera calibration parameter files and sample 2D pose data from `tests/data`

To run these scripts (and to use any of the functions data from S3), you need to set the `AWS_ACCESS_KEY_ID` and `AWS_SECRET_ACCESS_KEY` environment variables to keys belonging to an account that is a member of the Wildflower `cameras` group.


