本論文旨在建構高效能三維重建系統，以解決在大規模場景重建的龐大計算量問題。本系統透過整合最新的演算法實現，其主要步驟包含相機位置偵測、模型重建、網格簡化以及色彩映射，此外更提出基於相機位置的關鍵影像選取法，並將其應用在模型重建與色彩映射階段，以濾除多餘的影像資料以降低系統的運算量。實驗結果顯示本系統在數分鐘內完成大規模場景的重建且只造成微量的三維模型失真。相信此高效能的三維重建系統在未來將會有更多的延伸應用。

In this thesis, an efficient 3D reconstruction system is proposed to solve the problem of the huge computation of 3D reconstruction for large-scale scene. The system is developed by organizing the state-of-the-art algorithm, the main steps include camera pose estimation, modeling and simplification, and color mapping. Furthermore, a pose-based keyframe selection method is also proposed. It is applied to the stages of modeling and color mapping, which filters out the redundant data and leads to low computation. The results show that the proposed system only cost a few minutes for large-scale scene reconstruction with losing slight quality of 3D model. It is expected that the proposed system with high efficiency become more applied in the near future.

論文審定書 i
論文審定書(英文) ii
論文公開授權書 iii
致謝 iv
中文摘要 v
Abstract vi
Contents vii
List of Figures viii
List of Tables ix
Chapter 1 Introduction 1
1.1 Overview 1
1.2 Motivation 5
1.3 Contribution 6
1.4 Organization 7
Chapter 2 Background Review 8
2.1 SLAM System 8
2.2 Recent Literature of 3D Reconstruction 10
Chapter 3 System Development 13
3.1 Overview 13
3.1.1 Camera Pose Estimation 15
3.1.2 Modeling and Simplification 16
3.1.3 Color Mapping 17
3.2 Discussion of the Preliminary System 19
3.3 Keyframe Selection 22
3.3.1 Keyframe Selection for Color Mapping 23
3.3.2 Keyframe Selection for Modeling 26
Chapter 4 Experimental Results 30
4.1 Reconstructed Results of Fountain 31
4.2 Reconstructed Results of Sofa 35
4.3 Reconstructed Results of TUM_office 39
4.4 Reconstructed Results of Copyroom 43
4.5 Discussion 47
Chapter 5 Conclusions and Future Works 48
5.1 Conclusions 48
5.2 Future Works 49
Reference 50

[1] F. Endres, J. Hess, J. Sturm, D. Cremers, W. Burgard, "3D Mapping with an RGB-D Camera," IEEE Transactions on Robotics, 2014.
[2] T. Whelan, S. Leutenegger, R. F. S.-Moreno, B. Glocker and A. J. Davison "ElasticFusion: Dense SLAM without a pose graph." Robotics: Science and Systems, Vol. 11, 2015.
[3] S. Izadi, D. Kim, O. Hilliges, D. Molyneaux, R. Newcombe, P. Kohli, J. Shotton, S. Hodges, D. Freeman, A. Davison and A. Fitzgibbon, “KinectFusion: real-time 3D reconstruction and interaction using a moving depth camera,” in Proceedings of ACM Symposium on User Interface Software and Technology, pp.559-568, Oct. 2011.
[4] Q.-Y. Zhou and V. Koltun, “Dense scene reconstruction with points of interest,” ACM Transactions on Graphics, vol. 32, no. 4, July, 2013.
[5] Q.-Y. Zhou, S. Miller, and V. Koltun, “Elastic fragments for dense scene reconstruction,” in Proceedings of 2013 IEEE International Conference on Computer Vision, pp. 473-480, Dec., 2013.
[6] Q.-Y. Zhou and V. Koltun, “Simultaneous localization and calibration: self-calibration of consumer depth cameras,” in Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition, pp. 454-460, June 2014.
[7] S. Choi, Q.-Y. Zhou and V. Koltun, “Robust reconstruction of indoor scenes,” in Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition, pp. 5556-5565, Oct. 2015.
[8] M.-A. Raul, and J. D. Tardos, "ORB-SLAM2: an open-source SLAM system for monocular, stereo and RGB-D cameras," arXiv preprint arXiv:1610.06475, 2016.
[9] M.-A. Raul, J. M. M. Montiel and J. D. Tardos, "ORB-SLAM: A versatile and accurate monocular SLAM system," IEEE Transactions on Robotics, vol. 31, no. 5, pp. 1147-1163, Oct. 2015.
[10] B. Triggs, P. F. McLauchlan, R. I. Hartley and A. W. Fitzgibbon, "Bundle adjustment a modern synthesis," Vision Algorithms: Theory and Practice, pp. 298-372, 2000.
[11] E. Rublee, V. Rabaud, K. Konolige and G. Bradski, "ORB: An efficient alternative to SIFT or SURF," IEEE International Conference on Computer Vision, 2011.
[12] B. Curless and M. Levoy “A volumetric method for building complex models from range images,” SIGGRAPH, 1996.
[13] W. E. Lorensen, H. E. Cline, “Marching cubes: A high resolution 3D surface construction algorithm,” SIGGRAPH, 1987.
[14] C. Paolo, C. Marco, C. Massimiliano, D. Matteo, G. Fabio and R. Guido, "MeshLab: an open-source mesh processing tool, " in Proceedings of Eurographics Italian Chapter Conference, pp. 129-136, 2008.
[15] M. Waechter, N. Moehrle and M. Goesele, “Let there be color! – large-scale texturing of 3D reconstructions,” in Proceedings of European Conference on Computer Vision, pp. 836-850, 2014.
[16] Q.-Y. Zhou and V. Koltun, "Color map optimization for 3D reconstruction with consumer depth cameras," ACM Transactions on Graphics, 2014.
[17] J. Sturm, N. Engelhard, F. Endres, W. Burgard and D. Cremers, “A benchmark for the evaluation of RGB-D SLAM systems,” in Proceedings of 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, Dec. 2012.