Development Team

Mårten Björkman, Ph.D.
Lars Bretzner, Ph.D.
Pär Fornland, Ph.D.
Gareth Loy, Ph.D.

The InModo development team, consisting of researchers from an internationally leading research group in computer vision and image processing, has a very strong common competence base in this field. At the same time, the team members have specific complementary skills in low level machine oriented programming, data base handling, IT-project management, geometric transformations of importance for computer vision and real time programming. The team also has competence in human visual perception and the design of human-machine interfaces. Altogether, this gives the group a unique effectiveness and ability to realize InMoDo’s vision to make the digital camera a quick, user-friendly and extremely versatile tool for economically important large scale and mass applications.

The InMoDo development team is closely supported by Professor Jan-Olof Eklundh of the Computational Vision and Active Perception Laboratory at the Royal Institute of Technology in Stockholm (KTH), who has almost 40 years of experience in the vision field, and by Professor Robert Forchheimer of the Linköping Technical University (LiTH).

Hence, the team is extremely qualified to develop highly advanced and efficient state-of-the-art algorithms for:- visual recognition, forming the core of the applications- low level image handling- low level processing for image enhancement and distortion correction- intermediate level image segmentation and object detection- geometric correction based on advanced geometric theory and to integrate these into complete functioning camera phone based systems.

To be reliable and robust, as well as efficient from memory usage and computational points of view, such systems need to be based on precise modelling of the image formation and the scene contents, which is only becoming possible using the most advanced techniques developed in computer vision during the past few years. The InModo development team has deep and comprehensive insights into such techniques, some of which have been derived by the members themselves.

The team members are:

Mårten Björkman, Ph.D.
Mårten Björkman acquired his doctor’s degree in 2002 at KTH on a thesis in the field of computer vision, with specific focus on real time stereo and motion analysis for robots. He has many years of experience in low level machine oriented programming. Already in the early nineties he developed computer games for the Swedish TV, the first games in Sweden which allowed the television audience to play games directly via the telephone. In 1994, upon receiving his M.Sc. degree in computer technology, he worked three years as an application engineer for the company Mentor Graphics, with responsibility for software development for digital signal processing and VLSI design. For eight years he was active at the School for Computer Science and Communication at KTH, during the later years as a fellow in the research project CogVis directed towards the creation of systems for machine and artificial vision. Mårten Björkman has published scientific papers in the areas of computer vision, computer technology, artificial intelligence and robotics. To InMoDo he contributes competence in the areas of real time computer vision and low level machine oriented programming.

Lars Bretzner, Ph.D.
Lars Bretzner holds a M.Sc. in Eng. Physics from Lund University and a Ph.D. in Computer Science from the Computational Vision and Active Perception Laboratory (CVAP) at KTH. He has a broad industrial and academic experience in developing advanced real-time computer vision applications, e.g. novel human-computer interfaces based on computer vision and vision systems for automotive safety.

Pär Fornland, Ph.D.
Pär Fornland acquired his doctor’s degree in 1998 in computer vision on a dissertation regarding recognition of planar surfaces by the detection and analysis of patterns of motion. Already during his early college years Pär Fornland worked part time as a software developer for companies. His education in computer technology at the Linköping Technical University (LiTH) was completed by a thesis based on degree work in England. As a graduate student he published his research results at international conferences, was a tutor for several degree workers from foreign countries and a teacher for students of technology. He also represented KTH in Pan-European research projects, spent four months as a visiting research fellow in Italy and acted as a teacher and a freelance consultant in computer science. After graduation, he joined the Internet company Spray as a system’s architect, directing the development of Java-based web services used by several million unique daily users in some 14 European countries. Part of his work involved project management and coordination of international cooperation with departments in France, Germany and Armenia. During his affiliation with Spray, Pär Fornland was also in charge of a Developer Forum for enhancing the competence of the company’s developers. To InMoDo he contributes, besides competence in computer vision, experience from large scale commercial development work in project form, as well as skills in Java programming and data base handling in an environment of extensive web based systems.

Gareth Loy, Ph.D.
Gareth Loy completed his PhD at the Australian National University (ANU) in 2003, where his research focused on computer vision for human computer interaction. He also holds a Bachelor of Engineering with first-class honours (1999) and Bachelor of Science majoring in mathematics (1997) from ANU. During his doctoral studies he undertook consulting work as a research scientist for Seeing Machines Pty. Ltd., wrote an undergraduate lecture course in computer vision, and spent several months as a visiting researcher at the University of Western Australia and the Humanoid Interaction Lab at AIST, Japan. In 2003 he joined the Computer Vision and Active Perception Laboratory at KTH in Stockholm where he has worked on image and video analysis, and real-time feature detection and interpretation. He has also maintained his interest in commercial consulting, and established strong collaborative links between KTH, ANU and National ICT Australia (NICTA). He has published widely in international conferences and journals and authored several patents. To Inmodo he brings creative problem solving skills coupled with specialised knowledge in designing and prototyping real time computer vision algorithms.