Skip to main content

Tools for constructing 3D pose tracks from multi-camera 2D poses

Project description

poseconnect

Tools for constructing 3D pose tracks from multi-camera 2D poses

Installation

To install both the Python library and the command line interface, launch a Python virtual environment (optional, but recommended) and then install via pip:

pip install poseconnect

Command line interface

The command line interface consists of a single command (poseconnect) with four subcommands:

  • reconstruct takes a file of 2D pose data and a file of camera calibration info as inputs and outputs a file of 3D pose data
  • track takes a file of 3D pose data as input and outputs a file of 3D pose data with track IDs
  • interpolate takes a file of 3D pose data with track IDs as input and fills time gaps in each track with linearly interpolated poses
  • identify takes a file of 3D pose data with track IDs and a file of location sensor data with person IDs and outputs a file of 3D pose data with track and person IDs
  • overlay takes a file of 2D or 3D pose data and a source video for the data and outputs a video with the poses overlaid

Help

Overall help is available at poseconnect --help and help for each subcommand (including usage and a list of options) is available at poseconnect SUBCOMMAND --help (e.g., poseconnect reconstruct --help).

Data formats

More information about input and output data file formats is available here.

Demo

To see the command line interface in action, download and unzip the following sample JSON data files to your current directory and run the demo shell script:

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

poseconnect-0.17.2.tar.gz (37.7 kB view hashes)

Uploaded Source

Built Distribution

poseconnect-0.17.2-py3-none-any.whl (40.0 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page