High-quality texture reconstruction method for architectural painted patterns

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

Abstract

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

CLC Number:

TP391
TU851

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.

Figures/Tables 21

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

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

Fig. 3 Process of generating pattern descriptive files

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

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

Fig. 6 Pattern line draft generation process

Fig. 7 Minimum primitive extraction process

Fig. 8 Matching result graph

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

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

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

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

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

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

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

Fig. 14 Cloister wall layout

Fig. 15 Overview of model texture rendering

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

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

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

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])

References 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.