Generation of High-Resolution Mosaic for Photo-Realistic Texture-Mapping of Cultural Heritage 3D Models

The work investigates the problem of how information contained in different overlapping images of a scene can be combined to produce larger images of higher quality. The resulted images can be used for different applications like forensic image analysis, computer animation, special effects, 3D model texture mapping or panorama mosaic. In our case, high-resolution image mosaics of mural frescos are required for the texturing of a 3D model that will be used in a movie production. We developed a novel method for the derivation of a high quality mosaic using multi-resolution and multi-temporal images acquired from arbitrary positions and cameras. This method named 'constrained mesh-wise affine transformation' allows for seamless enhancement of the scene in the areas where higher resolution images are available. In this paper, we also discuss alternative procedures for the texture mapping of a 3D model using existing multi-resolution and multi-temporal imagery. The work has been done within a project aimed at a virtual and physical reconstruction of the destroyed Buddha statues of Bamiyan, Afghanistan.

[1]  Michal Irani,et al.  Improving resolution by image registration , 1991, CVGIP Graph. Model. Image Process..

[2]  Steve Mann,et al.  Virtual bellows: constructing high quality stills from video , 1994, Proceedings of 1st International Conference on Image Processing.

[3]  Richard Szeliski,et al.  Image mosaicing for tele-reality applications , 1994, Proceedings of 1994 IEEE Workshop on Applications of Computer Vision.

[4]  Wolfgang Niem,et al.  MAPPING TEXTURE FROM MULTIPLE CAMERA VIEWS ONTO 3D-OBJECT MODELS FOR COMPUTER ANIMATION , 1995 .

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

[6]  Mi-Suen Lee,et al.  View Synthesis from Unregistered 2-D Images , 1996, Graphics Interface.

[7]  Ming-Chao Chiang,et al.  Local blur estimation and super-resolution , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[8]  Michael Elad,et al.  Restoration of a single superresolution image from several blurred, noisy, and undersampled measured images , 1997, IEEE Trans. Image Process..

[9]  Andrew Zisserman,et al.  Automated mosaicing with super-resolution zoom , 1998, Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231).

[10]  Robert L. Stevenson,et al.  Spatial Resolution Enhancement of Low-Resolution Image Sequences A Comprehensive Review with Directions for Future Research , 1998 .

[11]  Linda G. Shapiro,et al.  Acquisition and visualization of colored 3D objects , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[12]  Takashi Fuse,et al.  A STUDY ON GEOMETRIC CORRECTION OF HISTORICAL MAPS , 1999 .

[13]  M. Irani,et al.  About Direct Methods , 1999, Workshop on Vision Algorithms.

[14]  Takeo Kanade,et al.  Limits on super-resolution and how to break them , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[15]  J. Niederoest.  Landscape as a historical object: 3D reconstruction and evaluation of a relief model from the 18th century , 2002 .

[16]  Sébastien Roy,et al.  Automatic relighting of overlapping textures of a 3D model , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[17]  Fabio Remondino,et al.  Photogrammetric Reconstruction of the Great Buddha of Bamiyan, Afghanistan , 2004 .

[18]  Marc Pollefeys,et al.  Radiometric alignment of image sequences , 2004, CVPR 2004.

[19]  M. Pollefeys,et al.  Radiometric Self-Alignment of Image Sequences ( CVPR ’ 04 ) , 2004 .