Computer vision research at the University of Oulu: from LBP to computational photography
Place: Large Lecture Room
Autors: Prof. Janne Heikkilä
Affiliation: Dep. of Computer Science and Engineering. Univ. of Oulu. Finland
Center for Machine Vision Research (CMV) at the University of Oulu (former Machine Vision Group) recently celebrated its 30th anniversary. Our long-term research on computer vision has resulted in a solid record of publications and merits both in fundamental and applied research. In particular, the Local Binary Patterns (LBP) method originating from the CMV has become probably the most commonly used texture operator within the computer vision community. Besides texture analysis, there are also many other areas of research at CMV including facial image analysis, geometric computer vision, and energy-efficient architectures for embedded systems.
In this talk, I will briefly present a general overview of the research activities at CMV and show some highlights of the recent results. The main emphasis is in my own research activities that include blur insensitive texture analysis using Local Phase Quantization (LPQ), geometric camera calibration, 3D reconstruction, and computational photography.