在這篇論文當中，我們能從多張數位影像自動重建出不確定平面架構的建築物平面模型。關於被建的場景中相關的位置資訊不用事先取得，必須先取得的只有相機的內部參數。
本三維立體模型重建系統包含三個主要步驟。首先我們需要取得兩兩影像的對應點與對應線。第二，場景中的半平面經由影像中的對應點與對應線計算出。最後，提出建構出完整平面模型的方法，包括半平面擴展的搜尋範圍限制，使用特徵點來進行平面擴展，將鄰近且共平面的兩平面組合起來，繪製出平面模型。我們已經使用這系統重建出數個立體盒模型與建築物模型，在論文中將會展示其中一些重建的結果。

In this paper, the automatic reconstruction of 3D uncertain plane models from a sequence of digital images of a real scene is proposed. No prior knowledge about the scene is needed to build the 3D plane models but the intrinsic camera parameters.
The 3D plane modeling task is decomposed into 3 main steps. First, the corresponding feature points and lines between images are extracted. Second, half-planes of the scene are computed by corresponding lines and points over multiple views. Final, the plane models producing method for complete plane regions is proposed involving limiting the search region of extending for a half-plane, extending region with feature points, grouping neighbor and coplanar planes and rendering the plane models. This system has been tested on several real boxes and buildings, and some of the reconstructed models are shown in this paper.

[1] R. Hartley, and A. Zisserman: Multiple View Geometry in Computer Vision. 2003.

[2] David G. Lowe: Distinctive Image Features from Scale-Invariant Keypoints. Computer Science Department, University of British Columbia. Jan, 2004.

[3] S. Heuel: Uncertain Projective Geometry: Statistical Reasoning for Polyhedral Object Reconstruction. 2004.

[4] Y-Z Wang, and P-C Chung: Reconstructing 3D Model of Real Scenes from Photographs. Institute of Computer and Communication National Cheng Kung University. Jul, 2005.

[5] C Baillard, C Schmid, A Zisserman, A Fitzgibbon: Automatic line matching and 3D reconstruction of buildings from multiple views. ISPRS Conference on Automatic Extraction of GIS Objects from Digital Imagery, IAPRS Vol. 32, Part 3-2W5, 1999.

[6] C Schmid, A Zisserman: Automatic line matching across views. Proc.Computer Vision and Pattern Recognition, 1997.

[7] S Heuel, W Forstner: Matching, reconstructing and grouping 3d lines from multiple views using uncertain projective Geometry. Proc. IEEE Conference on Computer Vision and Pattern, 2001.

[8] J Bauer, K Karner, K Schindler: Plane parameter estimation by edge set matching. Proc. 26th workshop of the Austrian Association for Pattern Recognition, Graz, Austria, 2002.

[9] K Schindler: Generalized Use of Homographies for Piecewise Planar Reconstruction. Scandinavian Conference on Image Analysis, Springer, 2003

[10] Open Source Computer Vision Library. http://www.intel.com/technology/computing/opencv/

[11] F Taillandier, R Deriche: Reconstruction of 3D linear primitives from multiple views for urban areas modelisation. PCV, 2002.

[12] Aerial views from Zisserman’s Data
http://www.robots.ox.ac.uk/~vgg/data/data-mview.html

[13] G Simon: Automatic Online Walls Detection for Immediate Use in AR Tasks. Mixed and Augmented Reality, 2006. ISMAR 2006. IEEE/ACM International Symposium on

[14] Y. Ma: An Invitation to 3-D Vision: From Images to Geometric Models. 2004 – Springer.

[15] Z. Zhang, R. Deriche, O. Faugeras, and Q. Luong: A Robust Technique for
Matching Two Uncalibrated Images through the Recovery of the Unknown Epipolar
Geometry. INRIA. 1994

[16] Z. Zhang: Flexible Camera Calibration by Viewing a Plane from Unknown Orientations. International Conference on Computer Vision (ICCV'99), Corfu, Greece, pages 666-673, September 1999.

[17] Z. Zhang: Token tracking in a cluttered scene. Image and Vision Computing, 1994

[18] S Coorg, S Teller: Extracting textured vertical facades from controlled close-range imagery. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1999.

[19] RT Collins: A space-sweep approach to true multi-image matching. Proc. Computer Vision and Pattern Recognition Conf, 1996.

[20] J Košecká, W Zhang: Extraction, matching, and pose recovery based on dominant rectangular structures, Computer Vision and Image Understanding, 2005.

[21] AY Yang, K Huang, S Rao, W Hong, Y Ma: Symmetry-based 3-D reconstruction from perspective images, Computer Vision and Image Understanding, 2005.

[22] M Lourakis, A Argyros, S Orphanoudakis: Detecting Planes In An Uncalibrated Image Pair, Proc. of BMVC’02, 2002.