3D Line Drawing for Archaeological Illustration

被引:0
作者
Tao Luo
Renju Li
Hongbin Zha
机构
[1] Peking University,Key Laboratory of Machine Perception (Ministry of Education)
来源
International Journal of Computer Vision | 2011年 / 94卷
关键词
Multi-scale 3D lines; Random walks; Minimum description length; Archaeological illustration;
D O I
暂无
中图分类号
学科分类号
摘要
Archaeological line drawing is an essential component of an archaeological report, which can not only illustrate heritage objects, but also record their accurate geometric measurements. In this paper, we propose a multi-scale approach to generating 3D line drawings on meshes reconstructed from raw scanning data of objects in grottoes. To reduce redundant lines detected on the rough and noisy surfaces, we first construct the discrete multi-scale representation of a given model based on random walks. The transition probability matrices are defined by studying the local variation around each vertex on the mesh. Furthermore, it is difficult to determine a single optimal scale for the whole mesh, because different scales are localized on the surface. A method for local scale selection is proposed based on the minimum description length (MDL) principle. Finally, we generate the line drawing using ridges or valleys detected with the selected scales. Experimental results show that the multi-scale 3D lines can well depict the shapes of heritage objects. Compared with the traditional manual method, our method is more accurate and convenient. Moreover, the offsets of detected lines are less than those using mesh smoothing methods. As assistance in archaeological mapping, our computer-generated line drawings can decrease the time cost to a large extent.
引用
收藏
页码:23 / 35
页数:12
相关论文
共 36 条
[1]  
DeCarlo D.(2003)Suggestive contours for conveying shape ACM Transactions on Graphics 22 848-855
[2]  
Finkelstein A.(2003)Bilateral mesh denoising ACM Transactions on Graphics 22 950-953
[3]  
Rusinkiewicz S.(1984)Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images IEEE Transactions on Pattern Analysis and Machine Intelligence 6 721-741
[4]  
Santella A.(2007)The Great Buddha Project: digitally archiving, restoring, and analyzing cultural heritage objects International Journal of Computer Vision 75 189-208
[5]  
Fleishman S.(2007)Apparent ridges for line drawing ACM Transactions on Graphics 26 19:1-19:7
[6]  
Drori I.(2008)Demarcating curves for shape illustration ACM Transactions on Graphics, SIGGRAPH Asia 27 157-116
[7]  
Cohen-Or D.(1998)Feature detection with automatic scale selection International Journal of Computer Vision 30 77-348
[8]  
Geman S.(2001)Gauss-Markov measure field models for low-level vision IEEE Transactions on Pattern Analysis and Machine Intelligence 23 337-612
[9]  
Geman D.(2004)Ridge-valley lines on meshes via implicit surface fitting ACM Transactions on Graphics 23 609-289
[10]  
Ikeuchi K.(2003)Multi-scale feature extraction on point-sampled models Computer Graphics Forum (Eurographics 2003) 22 281-431