本文發展一軟體套件名為 MB3DSOR 提供先驗建構與事後重建之功能。首先由俯視視角描述背景知識關於基於影像之被動式重建法；再探討單目與雙目攝影機之模型。由雙目攝影機模型探討基於紋理重建法名為運動恢復結構（Structure for motion）。由於運動恢復結構之缺點，本文亦探討體素（Voxel-based）重建法。體素重建法須以影像切割來進行剪影（Silhouette）提取；本文採用基於圖切割法（Graph-
cut based）與深度神經網路於不同條件下來影像切割。對本實驗室之開發光固化 3D 列印機，3DSPBM，先驗建構與事後重建可提供加工表現之評估功能。本論文末，探討先驗建構基於有限網格（Finite mesh）與體素（Voxel-based）重建法。

This thesis is devoted to developing a software package which provides solutions for these two purposes and it is named as MB3DSOR (Model-based 3D small object reconstruction package). It describes a bird-eye view of domain knowledge about image-based passive reconstruction problem at first. Then it discussed the primary model of cameras and stereo vision models. With these models, one can use the texture-matching method for reconstruction which is named as SFM (structure for motion). Due to some cons exist in the texture-based method, it discusses another method based on voxel. The voxel-based method needs image segmentation techniques for silhouette extraction. The classical graph-based and deep learning with neural networks are adopted for different conditions. For the applications in 3D stereolithography resin printer, a-prior 3D construction and reconstruction can be assisted in evaluating the manufacturing performance of the printer, especially Labs self-developing 3DP which is named as 3DSPBM (3D stereolithography printer based-on microscopy). Finally, it discusses the a prior 3D construct based on finite mesh and voxel-based reconstruction.

[1] Zoe Kleinman. Driverless car laser ruined camera. BBC News.Jun, 19, 2019. URL: https://www.bbc.com/news/technology-46875947
[2] Brad Templeton. Elon Musk’s War On LIDAR: Who Is Right And Why Do They Think That? Forbes News. May, 06, 2019. URL:
https://www.forbes.com/sites/bradtempleton/2019/05/06/elon-musks-war-on-lidar-who-is-right-and-why-do-they-think-that/#1023b1bd2a3b
[3] Silvio Savarese, Marco Andreetto, Holly Rushmeier, Fausto Bernardini, Pietro Perona. 3D Reconstruction by Shadow Carving: Theory and Practical Evaluation.
International Journal of Computer Vision, vol 71, issue 3, pp 305 ∼ 306. Springer Science and Business Media. 2006. doi: 10.1007/s11263-006-8323-9
[4] Adrian Kaehler, Gary Bradski. Learning OpenCV3. OReilly Media, Inc. 2016. ISBN : 1491937963, 9781491937969
[5] Richard J. Radke. Computer Vision For Visual EffectsCambridge University Press. 2013. ISBN : 0521766877, 9780521766876
[6] John G. Fryer, Duane C. Brown. Lens Distortion for Close-Range Photogrammetry. International Archives of Photogrammetry and Remote Sensing, vol 26 ,issue 5, pp 30∼37. 1986.
[7] Zhengyou Zhang. A Flexible New Technique for Camera Calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol 22, issue 11, pp 1330∼1334. 2000. doi: 10.1109/34.888718
[8] Richard Hartley, Andrew Zisserman. Multiple View Geometry in Computer Vision.Cambridge University Press. 2000. ISBN : 1139449141, 9781139449144
[9] Theo Moons, Maarten Vergauwen, Luc Van Gool. 3D Reconstruction From Multiple Images.Now Publishers Inc. 2009. ISBN: 1601982844, 9781601982841
[10] Yasutaka Furukawa, Carls Hernandez. Multi-View Stereo: A Tutorial.Now Publishers Inc. 2015. ISBN: 1601988362, 9781601988362
[11] Jianxiong Xiao. Multi-view 3D reconstruction for Dummies. Pricenton University. 2014.
[12] Styliani Verykokou, Charalabos Ioannidis. A Pothogrammetry-Based Structure From Motion Algorithm Using Robust Iterative Bundle Adjustment Techniques.ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information
Sciences. 2018.
[13] David G. Lowe. Distinctive Image Features from Scale-Invariant Keypoints. International Journal of Computer Vision, vol 50, issue 2, pp91∼110. 2004. doi: 10.1023/B:VISI.0000029664.99615.94
[14] Ives Rey-Otero, Mauricio Delbracio. Anatomy of the SIFT Method. Image Processing On Line Journal, vol 4, pp 370∼396. 2014. DOI: 10.5201/ipol.2014.82
[15] Chanop Silpa-Anan, Richard Hartley. Optimised KD-trees for fast image descriptor matching. IEEE Conference on Computer Vision and Pattern Recognition.
2008. doi: 10.1109/CVPR.2008.4587638
[16] Bruce Merry, James Gain, Patrick Marais. Accelerating kd-tree searched for all k-nearst neighbours. The Eurographics Association. 2013. doi :
10.2312/conf/EG2013/short/037-04
[17] Marius Muja, David G. Lowe. Fast Approximate Nearest Neighbors with Automatic Algorithm Configuration. International Conference on Computer Vision Theory
and Applications, pp 331∼340. 2009.
[18] Ben Ochoa. Lecture of CSE 252B: Computer Vision II. University of California San Diego. 2018.
[19] Sameer Agarwal, Noah Snavely, Steven M. Seitz, Richard Szeliski. Bundle Adjustment in the Large. European Conference on Computation Vision, pp 29 ∼ 42.
2010.
[20] Yaxiang Yuan. A Review of Trust Region Algorithms for Optimization. 1999. doi: 10.1.1.45.9964
[21] Manolis Lourakis, Antonis A. Argyros. Is Levenberg-Marquardt the Most Efficient Optimization Algorithm for Implementing Bundle Adjustment? IEEE International Conference on Computer Vision ,vol 1. 2005. doi: 10.1109/ICCV.2005.128
[22] Sam Roweis. Levenberg-Marquardt Optimization. New York University. URL:https://cs.nyu.edu/ roweis/notes/lm.pdf
[23] Jorge J. Mor´e. The Levenberg-Marquardt algorithm: Implementation and theory. Numerical Analysis, pp 105 ∼ 116. 1977.
[24] Kang Zhang, Yuqiang Fang, Dongbo Min, Lifeng Sun, Shiqiang Yang, Suicheng Yan, Qi Tian. Cross-Scale Cost Aggregation for Stereo Matching. IEEE Conference
on Computer Vision and Pattern Recognition. 2014. doi : 10.1109/CVPR.2014.206
[25] Ke Zhang, Jiangbo Lu, Gauthier Lafruit. Cross-based Local Stereo Matching Using Orthogonal Integral Images. IEEE Transactions on circuits and system for video
technology, vol 19, issue 7. 2009. doi : 10.1109/TCSVT.2009.2020478
[26] Maxime Lhuillier, Long Quan. Match Propagation for Image-Based Modeling and
Rendering. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol
24, issue 8, pp 1140 ∼ 1146. 2002. doi : 10.1109/TPAMI.2002.1023810
[27] Juho Kannala, Maekus Ylim¨

aki, Pekka Koskenkorva, Sami Sebastian Brandt.
Multi-View Surface Reconstruction by Quasi-Dense Wide Baseline Matching.
Emerging Topics in Computer Vision and its Application, pp 403 ∼ 424. 2011. doi
: 10.1142/9789814343008 0020
[28] Oxford University Visual Geometry Group. Link:
http://www.robots.ox.ac.uk/ vgg/data1.html
[29] Xinxin Zo, Chao Du, Sen Wang, Jiangbin Zheng, Ruigang Yang. Interative Visuall
Hull Refinement for Specular and Tranparent Object Surface Reconstruction. IEEE
International Conference on Computer Vision. 2015. doi : 10.1109/ICCV.2015.258
[30] Linus Fredriksson. Evaluation of 3D Reconstructing Based on Visual Hull Algo-
rithm. 2011.
[31] Svetlana Lazebnik, Yasutaka Furukawa, Jean Ponce. Projective Visual Hulls. In-
terational Journal of Computer Vision, vol 74, issue 2, pp 137 ∼ 165. 2007. doi :
10.1007/s11263-006-0008-x
[32] Li Guan, Sudipta Sinha, Jean-Sebastien Franco, Marc Pollefeys. Visual
Hull Construction in the Presence of Partial Occlusion. International Sym-
posium of 3D Data Processing, Visualization, and Transmission. 2006. doi:
10.1109/3DPVT.2006.147
[33] Aldo Laurentini. The Visual Hull Concept for Silhoutte-Based Image Understand-
ing. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol 16, issue
2, pp 150 ∼ 162. 1994. doi: 10.1109/34.273735
[34] Guido Gerig. Slide of CS6320 3D Computer Vision - Shape from Silhouette. New
York University. 2017.
[35] Kong-man Cheung, Simon Baker, Takeo Kanade. Shape-From-Silhouette Across
Time. International Journal of Computer Vision, vol 62, issue 3, pp 221 ∼ 247.
2005. doi : 10.1007/s11263-005-4881-5
[36] Maxim Mikhnevich, Patrick Hebert. Shape from Silhouette Under Varying Lighting
and Multi-viewpoints. Canadian Conference on Computer and Robotic Vision.
2011. doi : 10.1109/CRV.2011.45
[37] Pen Song, Xiaojun Wu, Michael Yu Wang. Volumetric stereo and silhouette fusion
for image-based modeling. The Visual Computer, vol 26, issue 12, pp 1435 ∼ 1450.
2010. doi : 10.1007/s00371-010-0429-y
[38] Adem Yasar Mulayim, Ulas Yilmaz, Volkan Atalay. Silhouette-based 3D Model
Reconstruction from Multiple Images. IEEE Transactions on Cybernetics, vol 33,issue 4, pp 582 ∼ 591. 2003. doi : 10.1109/TSMCB.2003.814303
[39] Kiriakos K. Kutulakos. A Theory of Shape by Space Carving. International
Journal of Computer Vision, vol 38, issue 3, pp 199 ∼ 218. 2000. doi :
10.1023/A:1008191222954
[40] Surya Prakash, Antonio Robles-Kelley. A semi-supervised approach to space
carving. International Conference on Pattern Recognition. 2009. doi :
10.1109/ICPR.2008.4761306
[41] Ping Chen, Min Dai, Kai Chen, Zhisheng Zhang. Rotation axis calibration of
a turntable using constrained global optimization. Optik, vol 125, issue 17, pp
4831 ∼ 4836. 2014. doi: 10.1016/j.ijleo.2014.04.047
[42] Andrew W. Fitzgibbon, Geoff Cross, Andrew Zisserman. Automatic 3D Model
Construction for Turn-Table Sequences. 3D Structure from Multiple Images of
Large-Scale Environments, pp 155 ∼ 170. 1998.
[43] Gabrial Taubin, Daniel Moreno, Douglas Lanman. 3D scanning for per-
sonal 3D printing: Build your own desktop 3D scanner. Siggraph. 2014. doi:
10.1145/2619195.2656314
[44] M. Tarini, M. Callieri, C. Montai, C. Rocchini. Marching Intersections: An Effi-
cient Approach to Shape-from-Silhoutte. Conference: Preoceedings of the Vision,
Modeling, and Visualization Conference. 2002.
[45] William E. Lorensen, Harvey E. Cline. Marching Cubes: A High Resolution 3D
Surface Construction Algorithm. SIGGRAPH annual conference on Computer
graphics and interactive techniques, vol 21, issue 4, pp 163 ∼ 169. 1987. doi:
10.1145/37402.37422
[46] Multi-View Stereo, Middlebury college, USA, 2006. URL:
http://vision.middlebury.edu/mview/
[47] Fr`edo Durand. Lenses and Depth of Field. Massachusetts Institute of Technology.
[48] Km. Pooja, Reghuandhan Rajesh. Image Segmentation: A Survey. International
Conference on Recent Advances in Mathematics, Statistics and Computer Science.
2015. doi : 10.1142/9789814704830 0049
[49] Song Yuheng, Yan Hao. Image Segmentation Algorithms Overview. 2017. arXiv:
1707.02051.
[50] Faliu Yi, Inkyu Moon. Image Segmentation: A Survey of Graph-cut methods.
International Conference on Systems and Informatics. 2012. doi : 10.1109/IC-
SAI.2012.6223428
[51] Yann LeCun, Patrick Haffner, L´eon Bottou, Yoshua Bengio. Object Recognition
with Gradient-Based Learning.Proceeding Shape, Contour and Grouping in Com-
puter Vision, Springer. UK. 1999. ISBN: 3-540-66722-9
[52] Alex Krizhevsky, Bya Sustskever, Geoffrey F. Hinton. ImageNet Classification
with Deep Convolutional Neural Networks. International Conference on Neural
Information Processing Systems, vol 1, pp 1097 ∼ 1105. 2012.
[53] Matthew D Zeilier, Rob Fergus. Visualizing and Understanding Convolutional Net-
works. 2013. arXiv: 1311.2901.
[54] Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir
Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich. Going Deeper
with Convolutions. 2014. arXiv: 1409.4842.
[55] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. Deep Residual Learning
for Image Recognition. 2015. arXiv: 1512.03385
[56] Gao Huang, Zhuang Liu, Laurens van der Maaten, Kilian Q. Weinberger. Densely
Connected Convolutional Networks. 2016. arXiv: 1608.06993
[57] Vincent Dumoulin, Francesco Visin. A guide to convolution arithmetic for deep
learning. 2016. arXiv: 1603.07285.
[58] G. Cybenko. Approximation by Superpositions of a Sigmoidal Function. Mathe-
matics of Control, Signals, and System, vol 2, issue 4, pp 303 ∼ 314. 1989. doi:
10.1007/BF02551274
[59] Kurt Hornik. Approximation capabilities of multilayer feedforward networks. Neu-
ral Networks, vol 4, issue 2, pp251 ∼ 257. 1991. doi: 10.1016/0893-6080(91)90009-
T
[60] Jonathan Long, Evan Shelhamer, Trevor Darrell. Fully Convolutional Networks
for Semantic Segmentation. 2014. arXiv: 1411.4038
[61] Olaf Ronneberger, Philipp Fischer, Thomas Brox. U-Net: Convolutional Networks
for Biomedical Image Segmentation. 2015. arXiv: 1505.04597
[62] Mo Zhang, Xiang Li, Mengjia Xu, Quanzheng Li. Image Segmentation and Classi-
fication for Sickle Cell Disease using Deformable U-Net. 2017. axXiv: 1710.08149.
[63] Vladimir Iglovikov, Alexey Shvets. TernausNet: U-Net with VGG11 Encoder Pre-
Trained on ImageNet for Image Segmentation. 2018. arXiv: 1801.05746.
[64] Paulo Jorge B´artolo. Stereolithography: Materials, Processes and Applications.
Springer, 2011. doi: 10.1007/978-0-387-92904-0
[65] Mark Venhse, Hermann Seitz. A New Micro-Stereolithography-System based on
Diode Laser Curing (DLC). International Journal of Precision Engineering And
Manufacturing, vol 15, issue 10, pp 2161 ∼ 2166. 2014. doi : 10.1007/s12541-014-
0577-5
[66] David T. Reid. Rapid Prototyping & Manufacturing Fundamentals of StereoLithog-
raphy. Society of Manufacturing Engineers. 1992. ISBN: 0070324336
[67] Edward Ford. Get to Know Your CNC: How to Read G-Code. Make:
Makezine. 2016. URL: https://makezine.com/2016/10/24/get-to-know-your-cnc-
how-to-read-g-code/