Work

Estimation of Virtual Moon Surface using LiDAR Measurements

Year: 2025

Type: article

Abstract: In this work, we present the MLS (Moon Lidar Simulation) system—a terrain height map and normal map prediction system for virtual lunar landscapes that uses LiDAR point cloud data as input. When operating the MLS system, we first simulate a virtual LiDAR sensor to sample height information from a 3D terrain and preprocess the data to generate an initial height map. We then leverage a Masked Autoencoder (MAE) structure to predict a normal map based on the height map input. Our system demonstrates its capability to predict occluded terrain areas, and we visualize the resulting height map and normal map in the Unity game engine.

Source: Journal of the Korea Computer Graphics Society

DOI: 10.15701/kcgs.2025.31.1.1

Authors: Mingi Yeom, Choongho Chung, Myeungun Kim, Daekwan Kim, Sung-Hee Lee

Institutions: Korea Advanced Institute of Science and Technology, Korea Aerospace Research Institute

Language: English

Topic: 3D Shape Modeling and Analysis

Subfield: Computational Mechanics

Field: Engineering

Domain: Physical Sciences

Open Access status: diamond

Funders: Korea Aerospace Research Institute, National Research Foundation of Korea

Awards: G5262869919, G7563628830