High-quality 3D reconstruction of weakly textured soft terrain based on 3D Gaussian splatting

Chuang Cao; Feiyu Zhang; Hongji Xia; Nan Luan

doi:10.1117/12.3052438

30 December 2024 High-quality 3D reconstruction of weakly textured soft terrain based on 3D Gaussian splatting

Chuang Cao, Feiyu Zhang, Hongji Xia, Nan Luan

Author Affiliations +

Proceedings Volume 13394, International Workshop on Automation, Control, and Communication Engineering (IWACCE 2024); 133940Z (2024) https://doi.org/10.1117/12.3052438
Event: International Workshop on Automation, Control, and Communication Engineering (IWACCE 2024), 2024, Hohhot, China

Abstract

3D map creation based on machine vision in a weak texture environment is a difficult problem of robot SLAM. Therefore, this paper tries a new method of low-texture map creation based on 3D Gaussian Splatting (3DGS). Through the acquisition of RGB images of the weak texture environment (lawn) by an Unmanned Aerial Vehicle (UAV) equipped with a monocular camera, the improved Structure from Motion (SfM) and 3DGS processing is used to generate a high-fidelity 3D map of complex and weak texture terrain. Experiments show that the peak signal-to-noise ratio (PSNR) of the training set is more than 40dB. For the new view, the Structural Similarity (SSIM) index in the frequency domain is improved by more than 0.7 in the low and mid-frequency range. While limited in high-frequency detail reconstruction, the method excels in global consistency and medium-scale feature capture, with efficient reconstruction and real-time rendering capabilities. This approach sets the stage for more advanced semantic mapping, promising applications in desert landscapes, barn interiors, and wilderness first aid scenarios.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Chuang Cao, Feiyu Zhang, Hongji Xia, and Nan Luan "High-quality 3D reconstruction of weakly textured soft terrain based on 3D Gaussian splatting", Proc. SPIE 13394, International Workshop on Automation, Control, and Communication Engineering (IWACCE 2024), 133940Z (30 December 2024); https://doi.org/10.1117/12.3052438

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Cameras

3D modeling

Global Positioning System

Atomic force microscopy

Visualization

3D acquisition

Data modeling

Keywords/Phrases

Search In:

Publication Years