Realistic 3D Reconstruction from Two Uncalibrated Views

Summary An approach for 3D surface reconstruction from two uncalibrated views is proposed in this paper. The SUSAN corner detector is used for corners from points in edges, which are detected by Sobel edge operator. After corners matching between two images, the fundamental matrix is estimated to acquire 3D structure from matched corners by an improved weighted linear algorithm, which is based on the epipolar geometry and the absolute conic theory. The triangular textures are acquired by dividing 2D image. Then a realistic 3D surface model is built by mapping the triangular textures to the 3D structure surface. Experiments show that the reconstructed 3D surface models are satisfactory and photo-realistic.

[1]  Carlo Tomasi.  Pictures and trails: a new framework for the computation of shape and motion from perspective image sequences , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Joseph O'Rourke,et al.  Computational Geometry in C. , 1995 .

[3]  O. Faugeras.  Stratification of three-dimensional vision: projective, affine, and metric representations , 1995 .

[4]  Jitendra Malik,et al.  Modeling and Rendering Architecture from Photographs: A hybrid geometry- and image-based approach , 1996, SIGGRAPH.

[5]  O. Faugeras,et al.  Camera Self-Calibration from Video Sequences: the Kruppa Equations Revisited , 1996 .

[6]  Jitendra Malik,et al.  Extracting Objects from Range and Radiance Images , 2001, IEEE Trans. Vis. Comput. Graph..

[7]  Peter Sturm,et al.  On focal length calibration from two views , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[8]  Chen Ze.  A Weighted Normalization Algorithm for Estimation of Fundamental Matrix , 2001 .

[9]  Paul A. Beardsley,et al.  Image-based 3D photography using opacity hulls , 2002, ACM Trans. Graph..

[10]  Stephen M. Smith,et al.  SUSAN—A New Approach to Low Level Image Processing , 1997, International Journal of Computer Vision.

[11]  Reinhard Koch,et al.  Visual Modeling with a Hand-Held Camera , 2004, International Journal of Computer Vision.

[12]  Peter Eisert,et al.  3D reconstruction of natural scenes with view-adaptive multi-texturing , 2004 .

[13]  Peter Eisert,et al.  3D reconstruction of natural scenes with view-adaptive multi-texturing , 2004, Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004..

[14]  Fernando Jaureguizar,et al.  Photorealistic 3D reconstruction from handheld cameras , 2005, Machine Vision and Applications.

[15]  Jong Seung Park,et al.  Interactive 3D reconstruction from multiple images: A primitive-based approach , 2005, Pattern Recognit. Lett..

[16]  Long Quan,et al.  A quasi-dense approach to surface reconstruction from uncalibrated images , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Alberto Del Bimbo,et al.  Metric 3D reconstruction and texture acquisition of surfaces of revolution from a single uncalibrated view , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Hui Zhang,et al.  Image-Based 3D Face Modeling System , 2005, EURASIP J. Adv. Signal Process..