New Czech-Israeli research grant at FIT

The objective of the new Czech-Israeli research co-operation is to find new methods for visual geo-location or estimation of depth from an image. The research grant was awarded to the team of Martin Čadík from FIT and the team of Professor Yosi Keller from Bar Ilan University. Together, the two teams will use the convolution neural networks to develop a new registration algorithm and descriptor of 2.5D and 3D models, a new method for localisation of cameras in the natural environment and a multimodal dataset for training networks predicting depth from a single image.

If the researchers manage to develop these new methods, the results of the joint research could advance the development of computational photography and could also be of some use in medicine or security.

The international project will last for three years. Its Czech part is financed by the Ministry of Education, Youth and Sports of the Czech Republic, the Israeli part by the Ministry of Science and Technology Israel.