Building a 3D point cloud using 360-degree video footage
Mattila, Sonja (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025052415567
https://urn.fi/URN:NBN:fi:amk-2025052415567
Tiivistelmä
The technology behind 3D point clouds and 360-degree video capturing is a developing area in many different industries. This thesis is research on how to produce a 3D point cloud and further a 3D model out of 360-degree video footage. Based on the research on theoretical framework and different testing in empirical part a manual was created for the use of students and teachers in Haaga-Helia 3D and robotics laboratory.
The first step was research from January to February about different 3D reconstruction technologies. There are multiple ways to create a 3D point cloud. All the technologies in theoretical framework and empirical part have a sensor which works as the key component when capturing the footage for the 3D point cloud.
After gaining enough information about 3D reconstruction the production of the manual started in March. The production started with filming suitable footage, extracting keyframes from it and using the keyframes to create a 3D point cloud with RealityCapture.
The manual was created based on the production steps in empirical part. When creating the manual certain steps were marked as optional ones based on success during production. The indicator for success of the production phase was a semi-realistic 3D model. The successful 3D model was created without splitting the keyframes which is why the step became optional in the manual.
The manual works as a base on how to create a 3D point cloud with 360-degree video technology. It is possible to further develop the manual if new software or hardware is acquired.
The first step was research from January to February about different 3D reconstruction technologies. There are multiple ways to create a 3D point cloud. All the technologies in theoretical framework and empirical part have a sensor which works as the key component when capturing the footage for the 3D point cloud.
After gaining enough information about 3D reconstruction the production of the manual started in March. The production started with filming suitable footage, extracting keyframes from it and using the keyframes to create a 3D point cloud with RealityCapture.
The manual was created based on the production steps in empirical part. When creating the manual certain steps were marked as optional ones based on success during production. The indicator for success of the production phase was a semi-realistic 3D model. The successful 3D model was created without splitting the keyframes which is why the step became optional in the manual.
The manual works as a base on how to create a 3D point cloud with 360-degree video technology. It is possible to further develop the manual if new software or hardware is acquired.