Earticle

초록

영어

The evaluation of Digital Elevation Model superresolution (DEM-SR) has predominantly relied on PSNR, often overlooking practical performance in physics-based VR simulations. Rather than proposing a new architecture, this work presents a quantitative case study evaluating existing Vision Transformers (SwinIR and Swin2SR) on real-world Copernicus GLO-30 terrain data. We assess their performance using both PSNR and application-centric metrics that capture statistical consistency and in-VR usability. Our analysis reveals that despite achieving respectable PSNR scores, architecturally advanced models can exhibit pronounced instability and immersion-breaking artifacts. We conclude that PSNR is insufficient in isolation and argue for incorporating practical, application-oriented metrics to ensure genuinely usable VR terrain.

목차

Abstract
I. INTRODUCTION
II. RELATED WORK
A. DEM Super-Resolution
B. The Limits of PSNR in Generative Tasks
C. Vision Transformers and Distributional Shift
III. EXPERIMENTAL SETUP
A. Data and Model
B. Evaluation Metrics
IV. RESULTS AND ANALYSIS
V. CONCLUSION AND FUTURE WORK
ACKNOWLEDGMENT
REFERENCES

키워드

저자

• Donghyun Woo [ Dept. of Advanced Defense Engineering Changwon National University Changwon, Republic of Korea ]
• GyuJin Shim [ School of Creative Convergence Education Handong Global University Pohang, Republic of Korea ]
• Sunjin Yu [ Meta-Convergence Content Major / Dept. of Artificial Intelligence Convergence Engineering Changwon National University Changwon, Republic of Korea ] Corresponding Author

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  한국차세대컴퓨팅학회 학술대회

• 간기

  반년간

• 수록기간

  2021~2025

• 십진분류

  KDC 566 DDC 004