Optimization and Evaluation of 3D Gaussian Splatting with Dynamic Opacity Filtering

doi:10.3969/j.issn.1006-2475.2025.11.011

Abstract

Abstract:
Abstract: 3D Gaussian splatting is an efficient modeling and rendering technique that demonstrates significant advantages in handling complex scenes and real-time rendering. However， when facing insufficient viewpoint coverage or complex lighting conditions， issues such as noise interference and inadequate 3D structure representation arise due to limited data support and algorithmic constraints， negatively impacting model accuracy and visual quality. To address these challenges， this paper presents an innovative 3D Gaussian iterative optimization method that introduces a filtering mechanism balancing density and opacity thresholds， significantly suppressing noise and enhancing the model’s geometric expressiveness in complex scenes. Furthermore， this paper presents a 3D structure evaluation system to assess the geometric accuracy and structural performance of Gaussian ellipsoids in 3D reconstruction. Experiments conducted on the DTU， Mip-NeRF 360， and Tanks & Temples datasets demonstrate that the proposed method not only maintains high visual quality in novel view generation but also surpasses existing baseline models in terms of 3D geometric accuracy. Additionally， it outperforms two state-of-the-art 3D Gaussian splatting-based mesh extraction methods: Surface-Guided Gaussian Reconstruction （SuGaR） and Gaussian Mesh Synthesis （GaMeS）. Through verification， the proposed filtering mechanism exhibits excellent adaptability across different scenes， effectively handling various complex lighting conditions and scene structures.

Key words: Key words: 3D Gaussian splatting； dynamic opacity threshold； noise suppression；3D stucture representation； novel view synthesis ,

CLC Number:

中图分类号：TP391

WAN Siyao1, FANG Zhiwei1, ZHOU Ting1, LYU Hanyun2, LUO Zheng3, CHEN Siyuan1. Optimization and Evaluation of 3D Gaussian Splatting with Dynamic Opacity Filtering[J]. Computer and Modernization, 2025, 0(11): 89-96.

References

［1］ LI X Y， ZHANG Q， KANG D， et al. Advances in 3D generation: A survey［J］. arXiv preprint arXiv:2401.17807， 2024.
［2］ WEGEN O， SCHEIBEL W， TRAPP M， et al. A Survey on non-photorealistic rendering approaches for point cloud visualization［J］. IEEE Transactions on Visualization and Computer Graphics， 2024，31（9）:4511-4533.
［3］ ZHANG J， WANG X， NI X D， et al. Neural radiance fields for multi-scale constraint-free 3D reconstruction and rendering in orchard scenes［J］. Computers and Electronics in Agriculture， 2024，217:108629.
［4］ KERBL B， KOPANAS G， LEIMKÜHLER T， et al. 3D Gaussian splatting for real-time radiance field rendering［J］. ACM Transactions on Graphics （ToG）， 2023，42（4）: 1-14.
［5］ FEI B， XU J Y， ZHANG R， et al. 3D Gaussian splatting as new era: A survey［J］. IEEE Transactions on Visualization and Computer Graphics， 2024，31（8）:4429-4449.
［6］ WANG H， LI M H. A new era of indoor scene reconstruction: A survey［J］. IEEE Access， 2024，12:110160-110192.
［7］ DARMON F， PORZI L， ROTA-BULÒ S， et al. Robust Gaussian splatting［J］. arXiv preprint arXiv:2404.04211， 2024.
［8］ SEISKARI O， YLILAMMI J， KAATRASALO V， et al. Gaussian splatting on the move: Blur and rolling shutter compensation for natural camera motion［J］. arXiv preprint arXiv:2403.13327， 2024.
［9］ CHEN W， LIU L G. Deblur-GS: 3D Gaussian splatting from camera motion blurred images［J］. Proceedings of the ACM on Computer Graphics and Interactive Techniques， 2024，7（1）:1-15.
［10］ GUÉDON A， LEPETIT V. SuGaR: Surface-aligned Gaussian splatting for efficient 3D mesh reconstruction and high-quality mesh rendering［C］// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE， 2024:5354-5363.
［11］ WACZYŃSKA J， BORYCKI P， TADEJA S， et al. GaMeS: Mesh-based adapting and modification of Gaussian splatting［J］. arXiv preprint arXiv:2402.01459， 2024.
［12］ LEE M Z， CHOU H S， LIU Y L， et al. 3D point cloud denoising algorithm based on two-stage filtering［C］// 2023 International Conference on Consumer Electronics-Taiwan （ICCE-Taiwan）. IEEE， 2023:419-420.
［13］ LEAL E， SANCHEZ-TORRES G， BRANCH J W. Sparse regularization-based approach for point cloud denoising and sharp features enhancement［J］. Sensors， 2020，20（11）: 3206.
［14］ HAQUE S M， GOVINDU V M. Robust feature-preserving denoising of 3D point clouds［C］// 2016 4th International Conference on 3D Vision （3DV）. IEEE， 2016:83-91.
［15］ CHEN S. An improved bilateral filtering based on curvature feature for point cloud denoising in road surface detection［C］// MIPPR 2023: Remote Sensing Image Processing， Geographic Information Systems， and Other Applications. SPIE， 2024，13088:152-158.
［16］ WATANABE R， NONAKA K， KATO H， et al. Point cloud denoising using normal vector-based graph wavelet shrinkage［C］// ICASSP 2022-2022 IEEE International Conference on Acoustics， Speech and Signal Processing （ICASSP）. IEEE， 2022:2569-2573.
［17］ WAHAB M F， GRITTI F， O'HAVER T C. Discrete Fourier transform techniques for noise reduction and digital enhancement of analytical signals［J］. TrAC Trends in Analytical Chemistry， 2021，143:116354.
［18］ WANG L C， CHEN Y J， SONG W H， et al. Point cloud denoising and feature preservation: An adaptive kernel approach based on local density and global statistics［J］. Sensors， 2024，24（6）:1718.
［19］ DUAN Y， YANG C C， LI H B. Low-complexity adaptive radius outlier removal filter based on PCA for LiDAR point cloud denoising［J］. Applied Optics， 2021，60（20）: E1-E7.
［20］ FISHER A， CANNIZZARO R， COCHRANE M， et al. ColMap: A memory-efficient occupancy grid mapping framework［J］. Robotics and Autonomous Systems， 2021，142: 103755.
［21］ GUO J L， FENG W X， HAO T F， et al. Denoising of a multi-station point cloud and 3D modeling accuracy for substation equipment based on statistical outlier removal［C］// 2020 IEEE 4th Conference on Energy Internet and Energy System Integration（EI2）. IEEE， 2020:2793-2797.
［22］ YAO Y， LUO Z X， LI S W， et al. MVSNet: Depth inference for unstructured multi-view stereo［C］// Proceedings of the European Conference on Computer Vision （ECCV）. Springer， 2018: 785-801.
［23］ OEHMCKE S， LI L， TREPEKLI K， et al. Deep point cloud regression for above-ground forest biomass estimation from airborne LiDAR［J］. Remote Sensing of Environment， 2024，302:113968.
［24］ JENSEN R， DAHL A， VOGIATZIS G， et al. Large scale multi-view stereopsis evaluation［C］// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE，2014:406-413.
［25］ BARRON J T， MILDENHALL B， VERBIN D， et al. MipNeRF 360: Unbounded anti-aliased neural radiance fields［C］// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE， 2022:5470-5479.
［26］ KNAPITSCH A， PARK J， ZHOU Q Y， et al. Tanks and temples: Benchmarking large-scale scene reconstruction［J］. ACM Transactions on Graphics （ToG）， 2017，36（4）: 1-13.
［27］ ZHANG R Y， ISOLA P， EFROS A A， et al. The unreasonable effectiveness of deep features as a perceptual metric［C］// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE， 2018:586-595.
［28］ ARABBOEV M， BEGMATOV S， RIKHSIVOEV M， et al. A comprehensive review of image super-resolution metrics: Classical and AI-based approaches［J］. Acta IMEKO， 2024，13（1）:1-8.

[1]	MEI Qin1, LIU Zhiren2, ZHAO Chenyang3, CHEN Yihan1, SU Mingrui3, WANG Gang3, WANG Junbo3. Real-time Secure Compression Method for Protection and Control Messages in 5G Power Distribution Networks [J]. Computer and Modernization, 2025, 0(11): 1-9.
[2]	XU Shengchao, LYU Junmin, JIANG Darui. Generative Artificial Intelligence Oriented Privacy Data Protection Method [J]. Computer and Modernization, 2025, 0(11): 10-16.
[3]	YANG Huimin, YANG Jin, SUN Yujie. Multi-strategy Dung Beetle Optimization Algorithm for Optimizing UAV Trajectory Planning Problem [J]. Computer and Modernization, 2025, 0(11): 16-31.
[4]	YOU Xindong, ZHAO Yuxian, LYU Xueqiang, LIU Boshan. Few-Shot Patent Classification Method Integrating Multi-dimensional Prompts and Multi-level Label Expansion [J]. Computer and Modernization, 2025, 0(11): 32-40.
[5]	ZHU Wenji1, BAN Weihua2, ZOU Lin3, LIU Xu3. Improved V-SLAM Method for Transmission Line Inspection UAV Based on LightGlue Network [J]. Computer and Modernization, 2025, 0(11): 41-48.
[6]	GUO Wangyong, HUANG Kun, LIANG Jiaben. Recognition of Composite Power Disturbances Based on Feature Selection and SCNGO-HKELM Algorithm [J]. Computer and Modernization, 2025, 0(11): 49-57.
[7]	WAN Chengkai1, AN Gaoyun2, CUI Lan3. Crowd Counting Estimation Algorithm of Railway Stations Based on Improved P2PNet [J]. Computer and Modernization, 2025, 0(11): 58-64.
[8]	SUN Erjie1, ZHANG Qifeng2, WANG Deqing3. Substation Equipment Defect Detection Based on Lightweight YOLOv8 [J]. Computer and Modernization, 2025, 0(11): 65-70.
[9]	WANG Jingpeng, CUI Yuyong, CAI Changlin, HE Ming’ao, LI Yinghao, TANG Zhonghe. Improved SOD Algorithm with Cross Modal Interaction and Multi Scale Aggregation [J]. Computer and Modernization, 2025, 0(11): 71-79.
[10]	YANG Anbo1, ZHONG Guoyun1, LIU Meifeng2, XI Chao2, ZHANG Wei1, DING Peng1. LSGI-YOLOv8: Ceramic Tile Surface Defect Detection Algorithm Based on Lightweight YOLOv8 [J]. Computer and Modernization, 2025, 0(11): 80-88.
[11]	QIU Xinyu1, CHEN Xiao1, CHEN Mingming2, GAO Hui3, MENG Ziyue3. Fault Diagnosis Method for Charging Pile Based on Vision Transformer [J]. Computer and Modernization, 2025, 0(11): 97-105.
[12]	XU Bo1, WEI Yijun1, ZENG Han2, KANG Futian3, HUANG Ying1, DENG Fangming2. Transformer Fault Diagnosis Method Based on Multi-source Data Fusion and Optimized Deep Belief Network [J]. Computer and Modernization, 2025, 0(11): 106-111.
[13]	LYU Hongjun, SHEN Lixin, LIU Lu, KANG Yipei. Dynamic Fleet Scheduling of Arctic Shipping Routes Under Beidou Third-generation Service [J]. Computer and Modernization, 2025, 0(11): 112-118.
[14]	ZHANG Hongtao, ZHU Yongzhong, ZHANG Yuxuan, XIA Shiyuan. Causal Inference Algorithm Driven by Denoising Diffusion Probabilistic Model [J]. Computer and Modernization, 2025, 0(11): 119-126.
[15]	LIU Longen1, SHI Dongxiang2, SHAN Baoming1, ZHANG Fangkun1, XU Qilei1. Lightweight Model for Recognizing Abnormal Behavior in Factory Personnel Based on Deep Learning [J]. Computer and Modernization, 2025, 0(10): 1-6.