视觉感知导向的实景三维建筑场景LOD自适应可视化

doi:10.11947/j.AGCS.2025.20240347

测绘学报 ›› 2025, Vol. 54 ›› Issue (6): 1054-1070.doi: 10.11947/j.AGCS.2025.20240347

视觉感知导向的实景三维建筑场景LOD自适应可视化

林浩嘉¹^,²^,³^,⁴(), 郭仁忠¹^,³^,⁴, 贺彪¹^,³^,⁴(), 蒯希¹^,³^,⁴, 马丁¹^,³^,⁴, 李程鹏¹^,³^,⁴

^1.深圳大学建筑与城市规划学院智慧城市研究院，广东　深圳　518060
^2.自然资源部数字制图与国土信息应用重点实验室，湖北　武汉　430079
^3.亚热带建筑与城市科学全国重点实验室，广东　深圳　518060
^4.粤港澳智慧城市联合实验室，广东　深圳　518060

收稿日期:2024-08-26 修回日期:2025-04-14 发布日期:2025-07-14
通讯作者: 贺彪 E-mail:linhaojia@szu.edu.cn;hebiao@szu.edu.cn
作者简介:林浩嘉（1993—），男，博士，研究方向为地图学、三维GIS、智慧城市等。E-mail：linhaojia@szu.edu.cn
基金资助:
国家自然科学基金(42401537);自然资源部数字制图与国土信息应用重点实验室开放研究基金(ZRZYBWD202304);深圳市科技重大专项(KJZD20230923115219040)

Visual-perception-oriented LOD adaptive visualization for realistic 3D building scenes

Haojia LIN¹^,²^,³^,⁴(), Renzhong GUO¹^,³^,⁴, Biao HE¹^,³^,⁴(), Xi KUAI¹^,³^,⁴, Ding MA¹^,³^,⁴, Chengpeng LI¹^,³^,⁴

^1.Research Institute for Smart City, School of Architecture and Urban Planning, Shenzhen University, Shenzhen 518060, China
^2.Key Laboratory of Digital Mapping and Land Information Application, Ministry of Natural Resources, Wuhan 430079, China
^3.State Key Laboratory of Subtropical Building and Urban Science, Shenzhen 518060, China
^4.Guangdong-Hong Kong-Macau Joint Laboratory for Smart Cities, Shenzhen 518060, China

Received:2024-08-26 Revised:2025-04-14 Published:2025-07-14
Contact: Biao HE E-mail:linhaojia@szu.edu.cn;hebiao@szu.edu.cn
About author:LIN Haojia (1993—), male, PhD, majors in cartography, 3D GIS, smart cities, et al. E-mail: linhaojia@szu.edu.cn
Supported by:
The National Natural Science Foundation of China(42401537);Open Research Fund Program of Key Laboratory of Digital Mapping and Land Information Application, Ministry of Natural Resources(ZRZYBWD202304);Shenzhen Major Science and Technology Special Projects(KJZD20230923115219040)

摘要/Abstract

摘要：

三维模型多分辨率表征（LOD）是三维场景可视化的重要技术。然而，对于具有真实程度高、尺度跨度大、数据规模大等特点的实景三维建筑场景可视化，仍存在可视化质量度量不准确、层次切换不连续、渲染效率不稳定等问题，难以在视点变换过程中保持视觉舒适性。本文提出了视觉感知导向的三维建筑场景LOD自适应可视化方法。首先，综合考虑三维渲染与视觉感知的过程与机理，对LOD可视化的视觉感知过程与特性进行归纳总结；然后，基于“三维空间-屏幕空间-视觉感知”的变换过程，对LOD可视化中简化模型误差的视觉感知进行量化表征，提出了基于视觉感知的LOD可视化质量度量方法；最后，将LOD可视化质量度量运用至LOD调度，综合考虑视觉关注和视点运动对可视化质量度量的影响，建立LOD层次选择准则，结合多线程调度技术，提出了顾及视觉感知特性的LOD自适应调度方法。试验结果表明，本文方法可以保持LOD切换过程的视觉舒适性，在可视化质量与效率之间保持动态平衡，实现了符合人眼视觉感知的三维建筑场景LOD自适应调度与渲染。

关键词: 视觉感知, LOD, 多尺度表达, 三维可视化, 实景三维

Abstract:

Multi-resolution representation of 3D models is an important technique for visualizing 3D scenes. However, when it comes to visualizing realistic 3D building scenes with high realism, large scale span, and large data volume, problems such as inaccurate visualization quality metrics, discontinuous level switching, and unstable rendering efficiency make it difficult to maintain visual comfort during viewpoint transformation. This study proposes a visual-perception-oriented LOD adaptive visualization method for realistic 3D building scenes. Firstly, the process of 3D rendering and the mechanism of visual perception are considered comprehensively, and the visual perception process and characteristics of LOD visualization are summarized. Secondly, based on the transformation process of “3D space-screen space-visual perception”, the visual perception of simplified model errors in LOD visualization is quantified, and a visual-perception-based quality metric for LOD visualization is proposed. Finally, an LOD selection criterion is established based on the LOD visualization quality metric and the impact of visual attention and viewpoint motion. Combined with multi-threaded scheduling technology, an LOD adaptive scheduling method that considers visual perception characteristics is proposed. Experimental results demonstrate that the proposed method effectively maintains visual comfort during LOD switching, ensuring a dynamic balance between visual quality and operating efficiency. Furthermore, the method facilitates adaptive scheduling and rendering of LOD for realistic 3D building scenes according to human visual perception.

Key words: visual perception, level of detail, multi-scale representation, 3D visualization, real-scene 3D

中图分类号:

P208

林浩嘉, 郭仁忠, 贺彪, 蒯希, 马丁, 李程鹏. 视觉感知导向的实景三维建筑场景LOD自适应可视化[J]. 测绘学报, 2025, 54(6): 1054-1070.

Haojia LIN, Renzhong GUO, Biao HE, Xi KUAI, Ding MA, Chengpeng LI. Visual-perception-oriented LOD adaptive visualization for realistic 3D building scenes[J]. Acta Geodaetica et Cartographica Sinica, 2025, 54(6): 1054-1070.

图/表 16

图1

图2

图3

图4

图5

图6

图7

图8

表1

图9

图10

表2

图11

表3

图12

表4

参考文献 44

[1]	郭仁忠, 陈业滨, 应申, 等. 三元空间下的泛地图可视化维度[J]. 武汉大学学报(信息科学版), 2018, 43(11): 1603-1610.
	GUO Renzhong, CHEN Yebin, YING Shen, et al. Geographic visualization of pan-map with the context of ternary spaces[J]. Geomatics and Information Science of Wuhan University, 2018, 43(11): 1603-1610.
[2]	张新长, 李少英, 周启鸣, 等. 建设数字孪生城市的逻辑与创新思考[J]. 测绘科学, 2021, 46(3): 147-152, 168.
	ZHANG Xinchang, LI Shaoying, ZHOU Qiming, et al. The rationale and innovative thinking of building digital twin city[J]. Science of Surveying and Mapping, 2021, 46(3): 147-152, 168.
[3]	朱庆, 张利国, 丁雨淋, 等. 从实景三维建模到数字孪生建模[J]. 测绘学报，2022, 51(6): 1040-1049. DOI: . doi: 10.11947/j.AGCS.2022.20210640
	ZHU Qing, ZHANG Liguo, DING Yulin, et al. From real 3D modeling to digital twin modeling[J]. Acta Geodaetica et Cartographica Sinica, 2022, 51(6): 1040-1049. DOI: . doi: 10.11947/j.AGCS.2022.20210640
[4]	WANG Ruisheng, HUANG Shangfeng, YANG Hongxin. Building3D: an urban-scale dataset and benchmarks for learning roof structures from point clouds[C]//Proceedings of 2023 IEEE/CVF International Conference on Computer Vision. Paris: IEEE, 2023: 20019-20029.
[5]	郭仁忠, 林浩嘉, 贺彪, 等. 面向智慧城市的GIS框架[J]. 武汉大学学报(信息科学版), 2020, 45(12): 1829-1835.
	GUO Renzhong, LIN Haojia, HE Biao, et al. GIS framework for smart cities[J]. Geomatics and Information Science of Wuhan University, 2020, 45(12): 1829-1835.
[6]	LIN Haojia, GUO Renzhong, MA Ding, et al. Digital-twin-based multi-scale simulation supports urban emergency management: a case study of urban epidemic transmission[J]. International Journal of Digital Earth, 2024, 17(1):2421950.
[7]	ZHANG Yunhao, ZHU Jun, ZHU Qing, et al. The construction of personalized virtual landslide disaster environments based on knowledge graphs and deep neural networks[J]. International Journal of Digital Earth, 2020, 13(12): 1637-1655.
[8]	YOU Jigang, ZHU Jun, LI Weilian, et al. ROI-constrained visualization of flood scenes to improve perception efficiency[J]. International Journal of Digital Earth, 2023, 16(1): 3065-3084.
[9]	CLARK J H. Hierarchical geometric models for visible surface algorithms[J]. Communications of the ACM, 1976, 19(10): 547-554.
[10]	MASON A E W, BLAKE E H. Automatic hierarchical level of detail optimization in computer animation[C]//Proceedings of 1997 Computer Graphics Forum. Oxford: Blackwell Publishers Ltd., 1997, 16(3): C191-C199.
[11]	朱庆, 龚俊, 杜志强, 等. 三维城市模型的多细节层次描述方法[J]. 武汉大学学报(信息科学版), 2005, 30(11): 965-969.
	ZHU Qing, GONG Jun, DU Zhiqiang, et al. LODs description of 3D city model[J]. Geomatics and Information Science of Wuhan University, 2005, 30(11): 965-969.
[12]	LUEBKE D, REDDY M, COHEN J D, et al. Level of detail for 3D graphics[M]. Amsterdam: Elsevier, 2002.
[13]	GIEGL M, WIMMER M. Unpopping: solving the image-space blend problem for smooth discrete LOD transitions[J]. Computer Graphics Forum, 2007, 26(1): 46-49.
[14]	KADA M, WICHMANN A, HERMES T. Smooth transformations between generalized 3D building models for visualization purposes[J]. Cartography and Geographic Information Science, 2015, 42(4): 306-314.
[15]	陈博, 佘江峰, 谈俊忠, 等. 三维场景中建筑物模型简化研究进展[J]. 武汉大学学报(信息科学版), 2020, 45(9): 1429-1437.
	CHEN Bo, SHE Jiangfeng, TAN Junzhong, et al. Research progress on simplification of building models in 3D scenes[J]. Geomatics and Information Science of Wuhan University, 2020, 45(9): 1429-1437.
[16]	AKENINE-MOLLER T, HAINES E, HOFFMAN N. Real-time rendering[M]. 4th ed. AK Peters/CRC Press, 2018.
[17]	CABRAL M, LEFEBVRE S, DACHSBACHER C, et al. Structure-preserving reshape for textured architectural scenes[J]. Computer Graphics Forum, 2009, 28(2): 469-480.
[18]	LIU Po, LI Chengming, LI Fei. Texture-cognition-based 3D building model generalization[J]. ISPRS International Journal of Geo-Information, 2017, 6(9): 260.
[19]	ZHANG Man, ZHANG Liqiang, TAKIS M P, et al. A geometry and texture coupled flexible generalization of urban building models[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2012, 70: 1-14.
[20]	SHE Jiangfeng, CHEN Bo, TAN Junzhong, et al. 3D building model simplification method considering both model mesh and building structure[J]. Transactions in GIS, 2022, 26(3): 1182-1203.
[21]	XIANG Hanyu, HUANG Xianfeng, LAN Feng, et al. A shape-preserving simplification method for urban building models[J]. ISPRS International Journal of Geo-Information, 2022, 11(11): 562.
[22]	杨必胜, 姜少波. 基于切割环分解的三维建筑物细节层次模型构造[J]. 测绘学报, 2011, 40(5): 575-581.
	YANG Bisheng, JIANG Shaobo. Generating levels of detail of 3D building models based on cutting loops decomposition[J]. Acta Geodaetica et Cartographica Sinica, 2011, 40(5): 575-581.
[23]	张春森, 张会, 郭丙轩, 等. 城市场景结构感知的网格模型简化算法[J]. 测绘学报，2020, 49(3): 334-342. DOI: . doi: 10.11947/j.AGCS.2020.20190068
	ZHANG Chunsen, ZHANG Hui, GUO Bingxuan, et al. Structure-aware simplified algorithm of mesh model for urban scene[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(3): 334-342. DOI: . doi: 10.11947/j.AGCS.2020.20190068
[24]	SHE Jiangfeng, GU Xiaoyan, TAN Junzhong, et al. An appearance-preserving simplification method for complex 3D building models[J]. Transactions in GIS, 2019, 23(2): 275-293.
[25]	CHEN Kan, JOHAN H, ERDT M. Appearance-driven conversion of polygon soup building models with level of detail control for 3D geospatial applications[J]. Advanced Engineering Informatics, 2020, 44: 101049.
[26]	程昫, 葛亮, 张帆, 等. 倾斜摄影建筑物模型自动纹理重建[J]. 测绘学报，2023, 52(9): 1528-1537. DOI: . doi: 10.11947/j.AGCS.2023.20220165
	CHENG Xu, GE Liang, ZHANG Fan, et al. Automated texture mapping of reality CSG building model with oblique aerial imagery[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(9): 1528-1537. DOI: . doi: 10.11947/j.AGCS.2023.20220165
[27]	BAI Haoyuan, SHEN Tao, HUO Liang, et al. Improved edge folding algorithm for 3D building models taking into account the visual features[J]. Buildings, 2023, 13(11): 2739.
[28]	殷智慧, 李朝奎, 吴柏燕, 等. 保持视觉特征的LoD模型简化算法研究[J]. 武汉大学学报(信息科学版), 2014, 39(7): 772-776.
	YIN Zhihui, LI Chaokui, WU Baiyan, et al. Simplification method for LoD model with the visual features preserved[J]. Geomatics and Information Science of Wuhan University, 2014, 39(7): 772-776.
[29]	WANG Chaoli, SHEN Hanwei. LOD map: a visual interface for navigating multiresolution volume visualization[J]. IEEE Transactions on Visualization and Computer Graphics, 2006, 12(5): 1029-1036.
[30]	CHE Li, KANG Fengju, HOU Xueli. Novel hierarchical level of detail combinatorial optimization method for large-scale 3D scene[J]. Journal of System Simulation, 2017, 29(9): 2073-2080.
[31]	FUNKHOUSER T A, SÉQUIN C H. Adaptive display algorithm for interactive frame rates during visualization of complex virtual environments[C]//Proceedings of the 20th Annual Conference on Computer Graphics and Interactive Techniques. Anaheim: ACM Press, 1993: 247-254.
[32]	HUANG Wumeng, CHEN Jing. A virtual globe-based time-critical adaptive visualization method for 3D city models[J]. International Journal of Digital Earth, 2018, 11(9): 939-955.
[33]	LI Jing, WU Huayi, YANG Chaowei, et al. Visualizing dynamic geosciences phenomena using an octree-based view-dependent LOD strategy within virtual globes[J]. Computers & Geosciences, 2011, 37(9): 1295-1302.
[34]	COZZI P, RING K. 3D engine design for virtual globes[M]. New York: AK Peters/CRC Press, 2011.
[35]	朱庆, 陈兴旺, 丁雨淋, 等. 视觉感知驱动的三维城市场景数据组织与调度方法[J]. 西南交通大学学报, 2017, 52(5): 869-876.
	ZHU Qing, CHEN Xingwang, DING Yulin, et al. Organization and scheduling method of 3D urban scene data driven by visual perception[J]. Journal of Southwest Jiaotong University, 2017, 52(5): 869-876.
[36]	徐海, 贺彪, 蒯希, 等. 融合分层层次细节和潜在可见集的三维城市场景高效可视化[J]. 测绘通报, 2024(1): 102-108.
	XU Hai, HE Biao, KUAI Xi, et al. Efficient visualization of 3D city scenes combining hierarchical level of details and potential visible sets[J]. Bulletin of Surveying and Mapping, 2024(1): 102-108.
[37]	NEHMÉ Y, DELANOY J, DUPONT F, et al. Textured mesh quality assessment: large-scale dataset and deep learning-based quality metric[J]. ACM Transactions on Graphics, 2023, 42(3): 1-20.
[38]	NEHMÉ Y, DUPONT F, FARRUGIA J P, et al. Visual quality of 3D meshes with diffuse colors in virtual reality: subjective and objective evaluation[J]. IEEE Transactions on Visualization and Computer Graphics, 2021, 27(3): 2202-2219.
[39]	KANDEL E R, SCHWARTZ J H, JESSELL T M, et al. (Eds.). Principles of neural science[M]. 5th ed. New York: McGraw-hill, 2013.
[40]	KASCHKE D M, DONNERHACKE D K, RILL M S. Optical devices in ophthalmology and optometry: technology, design principles, and clinical applications[M]. New York: John Wiley & Sons, 2014.
[41]	CARRASCO M. Visual attention: the past 25 years[J]. Vision Research, 2011, 51(13): 1484-1525.
[42]	CONNOR C E, EGETH H E, YANTIS S. Visual attention: bottom-up versus top-down[J]. Current Biology, 2004, 14(19): R850-R852.
[43]	RENNINGER L W, VERGHESE P, COUGHLAN J. Where to look next? eye movements reduce local uncertainty[J]. Journal of Vision, 2007, 7(3): 6.
[44]	BACHMANN T, FRANCIS G. Visual masking: studying perception, attention, and consciousness[M]. Oxford: Elsevier Academic Press, 2014.

LOD	M1			M2			M3			M4
LOD	TN	GE	PR	TN	GE	PR	TN	GE	PR	TN	GE	PR
0	4084	0	0.05	4325	0	0.04	9409	0	0.05	10607	0	0.05
1	2041	0.15	0.08	2161	0.18	0.09	4703	0.14	0.07	5302	0.20	0.10
2	1020	0.30	0.15	1080	0.36	0.18	2417	0.28	0.14	2651	0.40	0.20
3	780	0.61	0.30	539	0.72	0.36	1845	0.56	0.28	1793	0.79	0.40
4	596	1.21	0.61	447	1.44	0.72	1296	1.12	0.56	1204	1.59	0.79
5	417	2.42	1.21	374	2.87	1.44	885	2.25	1.12	817	3.17	1.59
6	298	4.84	2.42	267	5.75	2.87	577	4.49	2.25	412	6.34	3.17
7	198	9.69	4.84	132	11.49	5.75	333	8.99	4.49	207	12.69	6.34
8	104	19.37	9.69	66	22.98	11.49	168	17.98	8.99	103	25.37	12.69
9	—	—	—	—	—	—	83	35.96	17.98	—	—	—

视点	MN	TN	VN	FPS
远	80	425 201	903 219	171
中	180	420 860	804 008	152
近	98	161 287	255 213	163

状态	MN	TN	VN	FPS
关闭	210	528 127	876 602	155
开启	156	491 051	820 502	166

状态	MN	TN	VN	FPS
关闭	204	499 444	822 751	147
开启	175	405 524	737 903	164

视觉感知导向的实景三维建筑场景LOD自适应可视化

Visual-perception-oriented LOD adaptive visualization for realistic 3D building scenes

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 44

相关文章 15

编辑推荐

Metrics

本文评价

[1]	陈军, 田海波, 高崟, 张元杰, 刘万增, 武昊, 张宏伟, 黄蔚, 刘建军. 实景三维中国的总体架构与主体技术[J]. 测绘学报, 2025, 54(4): 636-649.
[2]	刘纪平, 周婷婷, 刘坡, 徐胜华, 王勇, 翟亮, 王琢璐, 祁俊杰, 马梦赫. 实景三维地理实体建模的思考与实践[J]. 测绘学报, 2025, 54(4): 650-660.
[3]	王庆栋, 王腾飞, 张力. 面向实景三维的点云跨模态对比掩码自编码预训练方法[J]. 测绘学报, 2025, 54(4): 675-687.
[4]	苏友能, 徐青, 孙群, 朱新铭, 张付兵, 刘波. 邻近边约束下的建筑物自动合并方法[J]. 测绘学报, 2025, 54(3): 563-576.
[5]	马威, 涂强, 潘建平, 赵立都, 涂伟, 李清泉. 桥梁实景三维高斯辐射场建模[J]. 测绘学报, 2024, 53(9): 1694-1705.
[6]	魏东, 刘欣怡, 张永军. 多视影像三维线云重建技术及其智能化发展展望[J]. 测绘学报, 2024, 53(6): 1025-1036.
[7]	肖腾, 王鑫, 梅熙, 叶志伟, 颜青松, 邓非. 摄影测量局部场景稳健合并的并行式运动恢复结构方法[J]. 测绘学报, 2024, 53(2): 332-343.
[8]	张荣庭, 张广运, 尹继豪. 复杂城市动态图卷积网络三维场景语义分割法[J]. 测绘学报, 2023, 52(10): 1703-1713.
[9]	艾廷华, 张翔. 地理信息科学中尺度概念的诠释与表达[J]. 测绘学报, 2022, 51(7): 1640-1652.
[10]	朱庆, 张利国, 丁雨淋, 胡翰, 葛旭明, 刘铭崴, 王玮. 从实景三维建模到数字孪生建模[J]. 测绘学报, 2022, 51(6): 1040-1049.
[11]	李清泉, 黄惠, 姜三, 胡庆武, 于文率. 优视摄影测量方法及精度分析[J]. 测绘学报, 2022, 51(6): 996-1007.
[12]	刘小波, 涂建光, 张海峰, 李治江. 国土综合整治三维模型碎片加载优化与调度[J]. 测绘学报, 2022, 51(4): 522-533.
[13]	行瑞星, 武芳, 巩现勇, 杜佳威, 刘呈熠. 建筑群组合直线模式识别的模板匹配方法[J]. 测绘学报, 2021, 50(6): 800-811.
[14]	刘鹏程, 肖天元, 肖佳, 艾廷华. 曲线多尺度表达的Head-Tail信息量分割法[J]. 测绘学报, 2020, 49(7): 921-933.
[15]	李德仁. 展望5G/6G时代的地球空间信息技术[J]. 测绘学报, 2019, 48(12): 1475-1481.