The existing 3D reconstruction methods cannot be effectively applied in architecture domain. A new method that is based on the architectural cartography criterion, techniques of AI and graphics recognition and understanding, is presented for automatic recognition and reconstruction of architectural structures by interpreting three-view drawings. In this method, automatic recognition of the architectural components?contours, integration of semantic information interpreted from symbols and notes etc., and analyzing topology relation of adjacent components are applied to reconstruct the whole building rapidly and effectually.Experimental results show that the method is efficient and accurate for automatic reconstruction of buildings composed of large number of components from their structure drawings with little user correction.
[1] Idesawa, M. A system to generate a solid figure from three views. JSME Bulletin, 1973,(16):216~225.
[2] Shin, B.-S., Shin, Y.G. Fast 3D solid model reconstruction from orthographic views. Computer-Aided Design, 1998,30(1):63~76.
[3] Gao, Wei, Peng, Qun-Sheng. 3D reconstruction based on 2D view feature. Chinese Journal of Computers, 1999,22(5):481~485 (in Chinese).
[4] Shum, S.S.P., Lau, W.S. Solid reconstruction from orthographic opaque views using incremental extrusion. Computer and Graphics, 1997,21(6):787~800.
[5] Dori, D. From engineering drawings to 3D CAD models: are we ready now? Computer-Aided Design, 1995,27(4):243~254.
[6] Dori, D., Weiss, M. A scheme for 3D object reconstruction from dimensioned orthographic views. Engineering Applications of Artificial Intelligence, 1996,9(1):53~64.
[7] Nezis, Konstantinos, Vosniakos, G. Recognizing 2.5D shape features using a neural network and heuristics. Computer-Aided Design, 1997,29(7):523~539.