Comparison of multi view stereo and neural radiance field for photogrammetric processing application / Muhammad Hariz Danial Mohd Radzi

Mohd Radzi, Muhammad Hariz Danial (2024) Comparison of multi view stereo and neural radiance field for photogrammetric processing application / Muhammad Hariz Danial Mohd Radzi. [Student Project] (Submitted)

Abstract

This research is about generating 3D models for topographic mapping purpose Multi-View Stereo (MVS) and Neural Radiance Field (NeRF). MVS faces challenges with reflective surfaces and under canopy, NeRF, shows that it can overcome these limitations. The study aims to evaluate MVS and NeRF in photogrammetric processing, with the objective of generating 3D models and assessing their quality. The MVS will generate the 3D model by using Agisoft Meta Shape software and NeRF will use NeRFstudio and both the 3D model will be compared using CloudCompare software which will overlap both model and we can identify the difference. The comparative analysis anticipates that NeRF will demonstrate reliability for topographic mapping, offering valuable insights for future applications. The expected outcomes include NeRF ability to predict under canopy, this shows that NeRF has become valuable tool for photogrammetric processing application. In summary, the research emphasizes NeRF's potential to enhance 3D modelling outcomes across diverse scenarios.

Metadata

Item Type: Student Project
Creators:
Creators
Email / ID Num.
Mohd Radzi, Muhammad Hariz Danial
UNSPECIFIED
Contributors:
Contribution
Name
Email / ID Num.
UNSPECIFIED
Mohd Salleh, Samsuri
UNSPECIFIED
Subjects: T Technology > TR Photography > Applied photography > Scientific and technical applications > Photogrammetry
Divisions: Universiti Teknologi MARA, Perlis > Arau Campus > Faculty of Architecture, Planning and Surveying
Programme: Bachelor of Surveying Science and Geomatics (Honours)
Keywords: multi view stereo, neural radiance field, photogrammetric
Date: July 2024
URI: https://ir.uitm.edu.my/id/eprint/105484
Edit Item
Edit Item

Download

[thumbnail of 105484.pdf] Text
105484.pdf

Download (198kB)

Digital Copy

Digital (fulltext) is available at:

Physical Copy

Physical status and holdings:
Item Status:

ID Number

105484

Indexing

Statistic

Statistic details