Skip to content

shashwatsuri/CalibSingleFromP2D

Repository files navigation

Methods for camera instrinsics and extrinsics (relative to ground plane) based on 2D pose.

Dependencies

Requires to run 2D detection, e.g., using the Pose2DSingle module

Install

The following script sets up a conda environment CalibSingleFrom2DP and installs required packages.

https://github.com/JunkyByte/easy_ViTPose

sh ./install.sh --easy_ViTPose

Run

The following script extracts frames and runs pose detection sh ./run.sh input_poses.json input_config.json output_folder

Input format

input_poses.json stores the 2D pose in the following json format: TODO:

input_config.json stores the index of head and ankle joints, TODO what else?

Output format

Output is the camera intrinsics, ground plane normal, ground plane position in the TODO .json format.

TODO: describe format in detail, we may have to code conversion functions to end up with a common format

Conventions and assumptions

People are up-right, if multiple people are present, their height is assumed to be the same.

The output calibration assumes a camera in right-handed coordinate system, i.e., x-axis points TODO right? y-axis points TODO up? z-axis points TODO forward/backwards?

About

This is a part of CasCalib by James Tang and me

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors