Point Cloud Fusion

Introduction

Point Cloud Fusion is a tool for fusing multiple point clouds into a unified, coherent model. This repository provides an easy-to-use interface for camera calibration and point cloud manipulation.

Getting Started

Prerequisites

Ensure you have Python 3.10 installed. The repository uses Open3D, which is compatible specifically with this Python version.

Installation Steps

1. Clone the GitHub repository:

   git clone git@github.com:voluview/fusion.git

2. Navigate into the cloned directory and install Poetry:

   cd fusion

3. Set up the virtual environment and install dependencies:

   poetry env use python3.10
   poetry install

   Note: Open3D requires Python 3.10; newer versions are not supported.

Configuration

You can customize the tool's behavior by modifying the settings in core/config.py. This file contains various parameters related to camera setup, point cloud alignment, and more.

How to Use

Camera Calibration

To achieve accurate point cloud fusion, you'll need to calibrate your cameras first.

Steps for Calibration:

1. Generate a ChaRuCo board: Utilize the utils/charuco_board.py script to generate a ChaRuCo board for calibration.

   make board

2. Run the calibration script:

   make cal

Controls:

Key	Action
q	Quit
c	Calibrate point clouds

Generating Point Clouds

After calibrating the cameras, proceed to generate the point clouds.

Steps to Generate Point Clouds:

1. Execute the point cloud script:

   make pcl

Controls:

Key	Action
q	Quit
a	Align point clouds with ICP
r	Reset alignment
s	Save point cloud alignment
d	Toggle depth view