面向建筑彩绘纹样的高质量贴图重构方法

doi:10.11996/JG.j.2095-302X.2024040804

图学学报 ›› 2024, Vol. 45 ›› Issue (4): 804-813.DOI: 10.11996/JG.j.2095-302X.2024040804

• 计算机图形学与虚拟现实 • 上一篇下一篇

面向建筑彩绘纹样的高质量贴图重构方法

龚辰晨¹(), 曹力¹^,²(), 张腾腾¹, 吴奕泽¹

1.合肥工业大学计算机与信息学院(人工智能学院)，安徽合肥 230009
2.合肥工业大学安全关键工业测控技术教育部工程研究中心，安徽合肥 230009

收稿日期:2024-01-18 接受日期:2024-04-17 出版日期:2024-08-31 发布日期:2024-09-03
通讯作者:曹力(1982-)，男，副教授，博士。主要研究方向为计算机辅助设计与计算几何。E-mail：lcao@hfut.edu.cn
第一作者:龚辰晨(1999-)，女，硕士研究生。主要研究方向为数字图像处理。E-mail：gongchenchen7@163.com
基金资助:
国家自然科学基金项目(61602146);国家自然科学基金项目(62272142);安徽省重点研究与开发项目(202104e11020006);青海省科技转化项目(2022-QY-203)

High-quality texture reconstruction method for architectural painted patterns

GONG Chenchen¹(), CAO Li¹^,²(), ZHANG Tengteng¹, WU Yize¹

1. School of Computer and Information (Artificial Intelligence College), Hefei University of Technology, Hefei Anhui 230009, China
2. Engineering Research Center of the Ministry of Education for Safety Critical Industrial Measurement and Control Technology, Hefei University of Technology, Hefei Anhui 230009, China

Received:2024-01-18 Accepted:2024-04-17 Published:2024-08-31 Online:2024-09-03
Contact: CAO Li (1982-), associate professor, Ph.D. His main research interests cover computer aided design and computational geometry. E-mail：lcao@hfut.edu.cn
First author：GONG Chenchen (1999-), master student. Her main research interest covers digital image processing. E-mail：gongchenchen7@163.com
Supported by:
National Natural Science Foundation of China(61602146);National Natural Science Foundation of China(62272142);Anhui Provincial Key Research and Development Project(202104e11020006);Qinghai Provincial Science and Technology Transformation Project(2022-QY-203)

摘要/Abstract

摘要：

建筑彩绘是绘制在木构建筑上的精美图案。在古建筑进行数字化展示时，通用处理方法是以网格模型加单张纹理的模式进行绘制。由于单张纹理贴图分辨率有限，无法展示所有细节，且常见的纹理为位图格式，使用多张高分辨率贴图会导致显存占用过大，致使数据交换效率变低。为解决上述难题，提出了一种高质量贴图重构方法。利用彩绘图案的自相似性和对称性，提取彩绘纹样最小不重复单元及版式信息。使用矢量数据表示最小图元并构建纹样素材库。在编辑三维模型的彩绘纹案时，通过复用图元并配置相应变换参数编码生成描述性文件，用以完成彩绘内容的渲染。实验结果表明，该方法有效减少了重复信息的存储，且提供更为清晰的细节，更好地进行数字化展示。

关键词: 真实感绘制, 图像矢量化表达, 版式结构, 纹理压缩, 建筑彩绘纹样

Abstract:

Architectural painted patterns refer to the exquisite patterns painted on wooden structures. When digitizing ancient architecture, the general solution involves using a mesh combined with a single texture map for rendering. However, due to the limited resolution of a single texture map, not all details can be adequately displayed. Moreover, common textures are stored pixel by pixel, and using multiple high-resolution texture maps can lead to excessive graphics processing unit memory usage, resulting in lower data exchange efficiency. To address the aforementioned challenges, a method for high-quality texture map reconstruction was proposed. This method employed the self-similarity and symmetry of the painted patterns, thereby extracting the smallest unduplicated pattern elements and layout information of the painted patterns. Vectorized data was utilized to represent the smallest pattern elements, and a library of pattern elements was constructed. When editing the painted patterns of a 3D model, these pattern elements were reused and configured with corresponding transformation parameters, which were encoded into descriptive files to complete the rendering of the painted content. Experimental results demonstrated that the proposed method could effectively reduce the storage of redundant information and provide a better presentation of details, thus enhancing realistic rendering.

Key words: realistic rendering, image vectorization representation, layout structure, texture compression, architectural painted patterns

中图分类号:

TP391
TU851

龚辰晨, 曹力, 张腾腾, 吴奕泽. 面向建筑彩绘纹样的高质量贴图重构方法[J]. 图学学报, 2024, 45(4): 804-813.

GONG Chenchen, CAO Li, ZHANG Tengteng, WU Yize. High-quality texture reconstruction method for architectural painted patterns[J]. Journal of Graphics, 2024, 45(4): 804-813.

图/表 21

图1 建筑彩绘原图((a)彩绘1；(b)彩绘2)

Fig. 1 Original architectural painting ((a) Painting 1; (b) Painting 2)

图2 建筑彩绘场景图((a)场景1；(b)场景2)

Fig. 2 Architecturalpaintedscene ((a) Scene 1; (b) Scene 2)

图3 纹样描述性文件加工流程

Fig. 3 Process of generating pattern descriptive files

图4 版式类型((a)四角中心型；(b)对角中心型；(c) 1/8型)

Fig. 4 Layout types ((a) Four-corner center type; (b) Diagonal center type; (c) 1/8 type)

图5 纹样线稿与模型纹样贴图的结构关系

Fig. 5 The structural relationship between pattern line drawing and model pattern map

图6 纹样线稿重构过程

Fig. 6 Pattern line draft generation process

图7 最小图元提取流程

Fig. 7 Minimum primitive extraction process

图8 匹配结果图

Fig. 8 Matching result graph

图9 变换信息及纹样线稿((a)变换信息；(b)矢量格式纹样线稿)

Fig. 9 Transformation information and pattern line draft ((a) Transformation information; (b) Vector format pattern line draft)

图10 素材库((a)通用图元；(b)几何图元；(c)植物图元；(d)自然图元)

Fig. 10 Material library ((a) General primitives; (b) Geometric primitives; (c) Plant primitives; (d) Natural primitives)

图11 纹样线稿图((a)文字纹；(b)卷纹；(c)植物纹)

Fig. 11 Line drawings of patterns ((a) Text pattern; (b) Scroll pattern; (c) Plant pattern)

图12 矢量格式的彩绘纹样贴图((a)文字纹；(b)卷纹；(c)植物纹)

Fig. 12 Painted pattern maps in vector format ((a) Text pattern; (b) Scroll pattern; (c) Plant pattern)

图13 根据构件类型分类的彩绘纹样贴图((a)，(b)大板类彩绘；(c)小构件类彩绘)

Fig. 13 Painted pattern maps categorized according to component type ((a), (b) Painted on large plates; (c) Painted on small components)

表1 彩绘纹样贴图与原图的结构相似性

Table 1 Structural similarity between painted pattern maps and original images

彩绘纹样类型	SSIM
文字纹	0.927 8
植物纹	0.861 6
卷纹	0.843 6

表2 彩绘纹样贴图与原图所占空间对比

Table 2 Comparison of the space occupied by the painted pattern map and the original image

彩绘纹样类型	原始图像占用空间/KB	彩绘纹样贴图占用空间/KB	压缩率/ %
文字纹	887	52	5.8
植物纹	887	121	13.6
卷纹	639	190	29.7

图14 回廊墙面布局图

Fig. 14 Cloister wall layout

图15 模型贴图渲染总览图

Fig. 15 Overview of model texture rendering

图16 不同视角观察下的回廊渲染结果((a)视角1；(b)视角2)

Fig. 16 Cloister render result from different perspectives ((a) Perspective 1; (b) Perspective 2)

图17 3种方法得到的纹理图对比

Fig. 17 Comparison of texture maps obtained by the three methods

表3 不同方法对比分析统计

Table 3 Comparative analysis statistics of different methods

方法	原始数据大小/MB	纹理预处理		平均渲染帧率/FPS
方法	原始数据大小/MB	纹理数据/ MB	减少数据存储百分比/%	平均渲染帧率/FPS
本文	132	17.16	87.0	58.7
文献[17]	132	49.76	62.3	51.0
文献[28]	132	36.62	72.3	57.0

图18 本文方法、文献[17]方法不同视距渲染效果((a)本文；(b)文献[17])

Fig. 18 The rendering effect of different viewing distances between the method of this article and the method of Reference 17 ((a) Ours; (b) Reference [17])

参考文献 28

[1]	JIANG Y M, YANG S, QIU H N, et al. Text2Human[J]. ACM Transactions on Graphics, 2022, 41(4): 1-11.
[2]	吴正存, 吴通, 敦晓荣. 基于可拓语义的澄城刺绣风格迁移设计方法研究[J]. 图学学报, 2023, 44(5): 1041-1049. DOI
	WU Z C, WU T, DUN X R. Research on design method of Chengcheng embroidery style migration based on extension semantics[J]. Journal of Graphics, 2023, 44(5): 1041-1049 (in Chinese).
[3]	何科雨, 陈中贵. 基于圆堆砌的纹理生成方法[J]. 图学学报, 2022, 43(6): 1114-1123.
	HE K Y, CHEN Z G. Circle packing based texture generation[J]. Journal of Graphics, 2022, 43(6): 1114-1123 (in Chinese). DOI
[4]	CHAKI J, DEY N. Texture Feature Extraction Techniques for Image Recognition[M]. Singapore: Springer Singapore, 2020: 67-81.
[5]	SHEN X, YANG J R, WEI C B, et al. DCT-mask: discrete cosine transform mask representation for instance segmentation[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 8716-8725.
[6]	YUAN X D, KORTYLEWSKI A, SUN Y H, et al. Robust instance segmentation through reasoning about multi-object occlusion[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 11136-11145.
[7]	DING H, QIAO S Y, YUILLE A, et al. Deeply shape-guided cascade for instance segmentation[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2021: 8274-8284.
[8]	ZHU H K, CAO J, XIAO Y Y, et al. TCB-spline-based image vectorization[J]. ACM Transactions on Graphics, 2022, 41(3): 1-17.
[9]	TODO H, KOBAYASHI K, KATSURAGI J, et al. Stroke transfer: example-based synthesis of animatable stroke styles[C]// Special Interest Group on Computer Graphics and Interactive Techniques Conference Proceedings. New York: ACM, 2022: 1-10.
[10]	SMITH S. Adaptive scalable texture compression[M]//GPU Pro 360 Guide to Mobile Devices. Boca Raton:Taylor & Francis, CRC Press, 2018: 153-166.
[11]	NAH J H. QuickETC2: fast ETC2 texture compression using Luma differences[J]. ACM Transactions on Graphics, 2020, 39(6): 270:1-270:10.
[12]	AKENINE-MÖLLER T, NILSSON J, ANDERSSON M, et al. Texture level of detail strategies for real-time ray tracing[M]// Ray Tracing Gems. Cham: Springer, 2019: 321-345.
[13]	ZHANG X Z, CHEN K, JOHAN H, et al. SLOD2+WIN: semantics-aware addition and LoD of 3D window details for LoD2 CityGML models with textures[J]. The Visual Computer, 2024, 40(1): 1-19.
[14]	ZHANG N X, ZHONG Y Z, DIAN S Y. Rethinking unsupervised texture defect detection using PCA[J]. Optics and Lasers in Engineering, 2023, 163: 107470.
[15]	姜展, 李梅, 孙振明, 等. 给予着色器的LOD纹理混合与反走样平滑过渡关键技术研究[J]. 北京大学学报: 自然科学版, 2022, 58(1): 113-122.
	JIANG Z, LI M, SUN Z M, et al. Research on key techniques of smoothing transition of LOD texture blending and antialiasing based on Shader[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2022, 58(1): 113-122 (in Chinese).
[16]	FIELDING E. Graphics texture mapping: US10825207[P]. 2020-11-03.
[17]	包馨, 周国清, 何宏昌. “分形-视距” (FVD)多分辨率纹理生成及组织方法[J]. 地球信息科学学报, 2021, 23(4): 551-563. DOI
	BAO X, ZHOU G Q, HE H C. Study on fractal viewing distance(FVD) multi-resolution texture data generation and organization[J]. Journal of Geo-Information Science, 2021, 23(4): 551-563 (in Chinese).
[18]	SURESH KUMAR R, MANIMEGALAI P. Near lossless image compression using parallel fractal texture identification[J]. Biomedical Signal Processing and Control, 2020, 58: 101862.
[19]	BAO X, ZHOU G Q, YUE T, et al. Research on 3d building visualization based on texture simplification and fractal compression[C]// International Conference on Geomatics in the Big Data Era. Hannover: The International Society for Photogrammetry and Remote Sensing, 2020: 179-185.
[20]	MULLA A, GUNJIKAR N, NAIK R R. Comparison of different image compression techniques[J]. International Journal of Computer Applications, 2013, 70: 7-12.
[21]	CHEN X, ZHANG S. Three dimensional range geometry and texture data compression with space-filling curves[J]. Optics Express, 2017, 25(21): 26103-26117. DOI PMID
[22]	SAYED Z, UGAIL H, PALMER I, et al. Auto-parameterized shape grammar for constructing Islamic geometric motif-based structures[M]// Transactions on Computational Science XXVIII. Heidelberg: Springer, 2016: 146-162.
[23]	赵海英, 王婷婷. 面向设计的装饰纹样基因生成算法研究[J]. 图学学报, 2019, 40(4): 810-815. DOI
	ZHAO H Y, WANG T T. Research on the design-oriented ornament pattern gene regeneration algorithm[J]. Journal of Graphics, 2019, 40(4): 810-815 (in Chinese).
[24]	赵海英, 解光鹏, 高子惠. 基于数学规则的龟背纹自动生成方法[J]. 北京邮电大学学报, 2022, 45(4): 110-115. DOI
	ZHAO H Y, XIE G P, GAO Z H. Automatic generation method of turtle back pattern based on mathematical rules[J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(4): 110-115 (in Chinese). DOI
[25]	TIAN G D, YUAN Q N, HU T, et al. Auto-generation system based on fractal geometry for batik pattern design[J]. Applied Sciences, 2019, 9(11): 2383.
[26]	李华飙, 侯小刚, 王婷婷, 等. 基于规则学习的传统纹样统一生成模式研究[J]. 浙江大学学报: 理学版, 2020, 47(6): 669-676, 680.
	LI H B, HOU X G, WANG T T, et al. An unified generation scheme of traditional patterns based on rule learning[J]. Journal of Zhejiang University: Science Edition, 2020, 47(6): 669-676, 680 (in Chinese).
[27]	SHI Q N, AN J C, GAGNON K K, et al. Image edge detection based on the canny edge and the ant colony optimization algorithm[C]// 2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics. New York: IEEE Press, 2019: 1-6.
[28]	REDDY P, GUERRERO P, MITRA N J. Search for concepts: Discovering visual concepts using direct optimization[EB/OL]. [2023-11-10]. https://arxiv.org/abs/2210.14808.

面向建筑彩绘纹样的高质量贴图重构方法

High-quality texture reconstruction method for architectural painted patterns

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 21

参考文献 28

相关文章 7

编辑推荐

Metrics

本文评价

[1]	柳有权 1，张彩荣 1，马雷 2,3，石剑 4，孙昭 1，陈彦云 3 . 一种基于图像结构特征的实时点画生成算法[J]. 图学学报, 2019, 40(3): 435-440.
[2]	邵绪强 1，刘艳 2，王新颖 1. 一种拉格朗日粒子流体的高效表面重建方法[J]. 图学学报, 2016, 37(5): 607-613.
[3]	钱文华，徐丹，普园媛，徐瑾，何磊，廖东伟. 非真实感绘制漫画艺术效果的研究与实现[J]. 图学学报, 2016, 37(2): 218-223.
[4]	喻扬涛，徐丹. 蜡染冰纹生成算法研究[J]. 图学学报, 2015, 36(2): 159-165.
[5]	李倩，解婉誉，李玉润，普园媛，徐丹. 云南绝版套刻版画的数字模拟合成技术研究[J]. 图学学报, 2013, 34(3): 120-125.
[6]	卢丽稳，普园媛，刘玉清，钱文华，徐丹. 云南重彩画人脸肖像生成算法[J]. 图学学报, 2013, 34(3): 126-133.
[7]	钱小燕. 基于图像的非真实感艺术绘制技术综述[J]. 图学学报, 2010, 31(1): 6-12.