Appendix 164 3. A 3. A. 1 Approximate Estimation of Fundamental Matrix from General Matrix 164 3. A. 2 Estimation of Affine Transformation 165 4 RECOVERY OF EPIPOLAR GEOMETRY FROM LINE SEGMENTS OR LINES 167 Line Segments or Straight Lines 168 4. 1 4. 2 Solving Motion Using Line Segments Between Two Views 173 4. 2. 1 Overlap of Two Corresponding Line Segments 173 Estimating Motion by Maximizing Overlap 175 4. 2. 2 Implementation Details 4. 2. 3 176 Reconstructing 3D Line Segments 4. 2. 4 179 4. 2. 5 Experimental Results 180 4. 2. 6 Discussions 192 4. 3 Determining Epipolar Geometry of Three Views 194 4. 3. 1 Trifocal Constraints for Point Matches 194 4. 3. 2 Trifocal Constraints for Line Correspondences 199 4. 3. 3 Linear Estimation of K, L, and M Using Points and Lines 200 4. 3. 4 Determining Camera Projection Matrices 201 4. 3. 5 Image Transfer 203 4. 4 Summary 204 5 REDEFINING STEREO, MOTION AND OBJECT RECOGNITION VIA EPIPOLAR GEOMETRY 205 5. 1 Conventional Approaches to Stereo, Motion and Object Recognition 205 5. 1. 1 Stereo 205 5. 1. 2 Motion 206 5. 1. 3 Object Recognition 207 5. 2 Correspondence in Stereo, Motion and Object Recognition as 1D Search 209 5. 2. 1 Stereo Matching 209 xi Contents 5. 2. 2 Motion Correspondence and Segmentation 209 5. 2. 3 3D Object Recognition and Localization 210 Disparity and Spatial Disparity Space 210 5.

Marker-based photogrammetric as-built reconstruction is a well established method to capture industrial sites. However, to reduce costs and to accelerate the reconstruction procedure, there is an increasing demand for marker-free reconstruction methods. Aiming to eliminate the need for markers, we present a novel approach to as-built reconstruction, which integrates technical drawings as orthographic images seamlessly into the photogrammetric reconstruction process. We introduce a new type of image, the so called co-registered orthographic and perspective (COP) image, which consists of an orthographic view integrated into a perspective view. Metric measurements and reconstruction are possible using only a single COP image. We discuss epipolar geometry for this type of image and we provide orientation and reconstruction methods custom tailored to COP images for points, lines, and cylindric objects. Multiple examples of application in industrial environment show the practical applicability of the presented methods and confirm that the requirements for accuracy in typical applications can well be met.

It is with great pleasure that I welcome you to Lake Tahoe for the 2005 Int- national Symposium on Visual Computing (ISVC). ISVC provides a common umbrella for the four main areas of visual computing: vision, graphics, visu- ization, and virtual reality. The goal of ISVC is to provide a common forum for researchers, scientists, engineers, and practitioners throughout the world to present their latest research ?ndings, ideas, developments, and applications in the broader area of visual computing. The program consists of six oral sessions, two poster sessions, seven special tracks,fourkeynotepresentations,andoneinvitedpresentation.Theresponseto thecallforpapersforthegeneralISVC2005sessionswasverygood.Wereceived over110submissionsfromwhichweaccepted33papersfororalpresentationand 26 papers for poster presentation. Special track papers were solicited separately through the organizing and program committees of each track. A total of 32 papers were accepted for inclusion in the special tracks. All papers were reviewed with an emphasis on their potential to contribute to the state of the art in the ?eld. Selection criteria included accuracy and originality of ideas, clarity and signi?cance of results, and presentation qu- ity. The review process was quite rigorous, involving two or three independent double-blind reviews followed by a one-week discussion period. During the d- cussion period we tried to correct anomalies and errors that might have existed in the initial reviews. Despite our e?orts, we recognize that some papers worthy of inclusion may not have been included in the program. We o?er our sincere apologies to authors whose contributions might have been overlooked. IwishtothankeverybodywhosubmittedtheirworktoISVC2005forreview.

In the four years of its existence, MICCAI has developed into the premier - nual conference on medical image computing and computer-assisted interv- tion. The single-track conference has an interdisciplinary character, bringing - getherresearchersfromboththenaturalsciencesandvariousmedicaldisciplines. It provides the international forum for developments concerning all aspects of medical image processing and visualization, image-guided and computer-aided techniques, and robot technology in medicine. The strong interest in MICCAI is con?rmed by the large number of subm- sions we received this year, which by far surpassed our expectations. The arrival of the shipload of papers just before the deadlines (one in the European and the otherin theAmericantime zone)wasa particularlyenjoyableexperience,aswas the whole procedure of preparing the scienti?c programme. Both the quantity and quality of the submissions allowed us to compose a volume of high quality papers, which we are sure will contribute to the further development of this exciting ?eld of research. As for the hard numbers, in total 338 submissions were received. Next to full papers, short communications were solicited for works in progress,hardware prototypes, and clinical case studies. Long papers were reviewed by three or four reviewers and short papers by two or three reviewers. The ?nal selection of papers was carried out by the Programme Board. Out of the 246 long papers, 36 were accepted for oral presentation and 100 as full posters. An additional 75 of the long papers, and 47 out of 92 short papers were accepted as short posters.

This text presents techniques for describing image textures. Contrary to the usual practice of embedding the images to known modelling frameworks borrowed from statistical physics or other domains, this book deduces the Gibbs models from basic image features and tailors the modelling framework to the images. This approach results in more general Gibbs models than can be either Markovian or non-Markovian and possess arbitrary interaction structures and strengths. The book presents computationally feasible algorithms for parameter estimation and image simulation and demonstrates their abilities and limitations by numerous experimental results.