摘要: In this paper, we describe a novel method to create the complete 3D model of object on uncalibrated images. First, match points both detected by multi- scale Harris corner detection algorithm and line technique. Second, perform projected reconstruction based factorization using Singular Value Decomposition (SVD). After that, are able upgrade from projective Euclidean structure then eliminate ambiguity in reconstruction. Finally, use registration common build whole object. Sufficient experiments proved validity efficiency method.
thinkmind.org参考文章(6)
Toshio Ueshiba, Fumiaki Tomita, A Factorization Method for Projective and Euclidean Reconstruction from Multiple Perspective Views via Iterative Depth Estimation european conference on computer vision. pp. 296- 310 ,(1998) , 10.1007/BFB0055674
Jiangjian Xiao, Shah, Two-frame wide baseline matching international conference on computer vision. ,vol. 2, pp. 603- 609 ,(2003) , 10.1109/ICCV.2003.1238403
Thuy Tuong Nguyen, Xuan Dai Pham, Jae Wook Jeon, None, An improvement of the Standard Hough Transform to detect line segments international conference on industrial technology. pp. 1- 6 ,(2008) , 10.1109/ICIT.2008.4608701
David G. Lowe, Distinctive Image Features from Scale-Invariant Keypoints International Journal of Computer Vision. ,vol. 60, pp. 91- 110 ,(2004) , 10.1023/B:VISI.0000029664.99615.94
Zhiyong Ye, Yijian Pei, Jihong Shi, An Adaptive Algorithm for Harris Corner Detection computational intelligence. pp. 1- 4 ,(2009) , 10.1109/CISE.2009.5366231
Krystian Mikolajczyk, Cordelia Schmid, Scale & Affine Invariant Interest Point Detectors International Journal of Computer Vision. ,vol. 60, pp. 63- 86 ,(2004) , 10.1023/B:VISI.0000027790.02288.F2