@article{ART003109543},
author={Yeong-In Lee and Jin-Nyeong Heo and Ji-Hwan Moon and Ha-Young Kim},
title={Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2024},
volume={29},
number={8},
pages={23-32}
TY - JOUR
AU - Yeong-In Lee
AU - Jin-Nyeong Heo
AU - Ji-Hwan Moon
AU - Ha-Young Kim
TI - Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement
JO - Journal of The Korea Society of Computer and Information
PY - 2024
VL - 29
IS - 8
PB - The Korean Society Of Computer And Information
SP - 23
EP - 32
SN - 1598-849X
AB - NVS (Novel View Synthesis) is a field in computer vision that reconstructs new views of a scene from a set of input views. Real-time rendering and high performance are essential for NVS technology to be effectively utilized in various applications. Recently, 3D-GS (3D Gaussian Splatting) has gained popularity due to its faster training and inference times compared to those of NeRF (Neural Radiance Fields)-based methodologies. However, since 3D-GS reconstructs a 3D (Three-Dimensional) scene by splitting and cloning (Density Control) Gaussian points, the number of Gaussian points continuously increases, causing the model to become heavier as training progresses. To address this issue, we propose two methodologies: 1) Gaussian blending, an improved density control methodology that removes unnecessary Gaussian points, and 2) a performance enhancement methodology using a depth estimation model to minimize the loss in representation caused by the blending of Gaussian points. Experiments on the Tanks and Temples Dataset show that the proposed methodologies reduce the number of Gaussian points by up to 4% while maintaining performance.
KW - Novel View Synthesis;3D Gaussian Splatting;Computer Vision;NeRF;Light-Weighting;Deep Learning
DO -
UR -
ER -
Yeong-In Lee, Jin-Nyeong Heo, Ji-Hwan Moon and Ha-Young Kim. (2024). Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement. Journal of The Korea Society of Computer and Information, 29(8), 23-32.
Yeong-In Lee, Jin-Nyeong Heo, Ji-Hwan Moon and Ha-Young Kim. 2024, "Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement", Journal of The Korea Society of Computer and Information, vol.29, no.8 pp.23-32.
Yeong-In Lee, Jin-Nyeong Heo, Ji-Hwan Moon, Ha-Young Kim "Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement" Journal of The Korea Society of Computer and Information 29.8 pp.23-32 (2024) : 23.
Yeong-In Lee, Jin-Nyeong Heo, Ji-Hwan Moon, Ha-Young Kim. Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement. 2024; 29(8), 23-32.
Yeong-In Lee, Jin-Nyeong Heo, Ji-Hwan Moon and Ha-Young Kim. "Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement" Journal of The Korea Society of Computer and Information 29, no.8 (2024) : 23-32.
Yeong-In Lee; Jin-Nyeong Heo; Ji-Hwan Moon; Ha-Young Kim. Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement. Journal of The Korea Society of Computer and Information, 29(8), 23-32.
Yeong-In Lee; Jin-Nyeong Heo; Ji-Hwan Moon; Ha-Young Kim. Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement. Journal of The Korea Society of Computer and Information. 2024; 29(8) 23-32.
Yeong-In Lee, Jin-Nyeong Heo, Ji-Hwan Moon, Ha-Young Kim. Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement. 2024; 29(8), 23-32.
Yeong-In Lee, Jin-Nyeong Heo, Ji-Hwan Moon and Ha-Young Kim. "Gaussian Blending: Improved 3D Gaussian Splatting for Model Light-Weighting and Deep Learning-Based Performance Enhancement" Journal of The Korea Society of Computer and Information 29, no.8 (2024) : 23-32.