Application of complete affine invariants and pose determination for 3d intensity data, in geometric problems of 3d computer vision

Lines connect the coordinator with its partners. CORDIS does not always give exact coordinates for projects before 2014. These points are placed at city or country level.

Project objective

We deal with the problem of designing algorithms for automatic recognition of threedimensional objects by digital machines. Our approach is the formation of numerical features which remain invariant with respect to viewpoint changes. Within a HCM fellowship recently carried out, we could improve in several directions existing results for the special case of 3D intensity images (completeness, independence and orthogonality of the invariants). We propose applying these results to geometric problems of 3D Computer Vision. It is expected that our approach will minimize the requirements for projection views and search for corresponding points in several instances of geometric 3D Computer Vision.

Original text from CORDIS.

Participants

• Technische Universität Hamburg-Harburg · HamburgCoordinatorCity levelGermany
• Not availableCity levelGreece

Links

• View on CORDIS