Scale Space and Variational Methods in Computer Vision

Preface Image processing, computation, robot and machine vision are terms that refer to automatic visual perception through intelligent processing of image content. Such a demand requires the development of appropriate mathematical models which reformulate the answer to the perception problem as the lowest potential of a specifically designed objective function. The development of such models capable of reproducing human vision is a long shot objective in the domain. Variational methods are a very popular selection to address a number of components of visual perception, while scale space methods introduce the notion of hierarchical representation of image content, or property often present in biological autonomous perception organisms. The First International Conference on Scale Space and Variational Methods in Computer Vision (SSVM 2007) was an attempt to bring together two different communities with adjacent research interests, the one of scale-space analysis and the one of variational, geometric and level set (VLSM). Such a conference was a joint edition of the 4th VLSM and 6th Scale Space with aim to bring together various disciplines working in the area of visual perception (mathematicians, physicists, computer science, computational science, etc.). It gathered the attention of an important international scientific crowd with submissions and presentations from approximately twenty countries (Austria, Australia, Belgium,Canada, Switzerland, China, Germany, Denmark, Spain, France, Greece, Honk Kong, Israel, India, Ireland, Italy, Japan, Korea, Mexico, The Netherlands, Norway, Poland, Sweden, Turkey, England, USA) from the leading scientists from the domain. We received 133 high-quality full paper double-blind submissions. Each paper was reviewed by at least three members of the Program Committee. These reviews were considered from the Area Chairs who finally proposed 79 to be accepted. We selected 24 manuscripts for an oral presentation and 55 for poster presentation. Both oral and poster papers attributed the same length of pages in the conference proceedings. Furthermore we invited keynote speakers who can provide valuable additional inspirations beyond the mainstream topics in scale-space analysis and variational methods. It was our pleaser to welcome Prof. Franco Brezzi of University of Pavia, Institute for Advanced Study and IMATI-CNR (Italy), Prof. Emmanuel Candes of California Institute of Technology, (USA) and Prof. Peter Schroder of California Institute of Technology, (USA) as keynote speakers. We would like to thank the authors for their contributions, and the members of the Program Committee for their time and valuable comments during the review process. We would like also to acknowledge the support of Christian Trocchi, Daniela Casaburi and Livia Marcellino for their help with the web-site and organization. Last but not least special thanks to Francesca Incensi for handling the submission/review/decisions and proceedings aspects of the conference. Finally we are grateful to the University of Bologna, the University of Naples Federico II, GNCS-INDAM, CINECA Bologna and CIRAM (Research Centre of Applied Mathematics) Bologna for their sponsorship. It is our belief that this conference will become a reference in the domain, and will contribute on the development of new ideas in the area of visual perception through processing images with mathematical models. May-June 2007, Fiorella Sgallari , Almerico Murli, Nikos Paragios