911������

A computer can recognize, classify and analyze items in the images automatically and react on events with a help of a camera. It is even possible to analyze large amounts of images and look for certain types of phenomena, which would be difficult to perceive and interpret solely by humans.

- 3D-modeling may concern for instance autonomous moving of robots, such as self-driving cars. Perceiving other moving objects on the street is then crucial, yet problems may arise due to varying conditions, clothing and types of transportation vehicles, states Kannala.

Doctoral student Iaroslav Melekhov follows his supervisor Kannala to Aalto from the University of Oulu. During the spring Kannala is going to recruit more Computer Vision researchers to his group.

Academy of Finland project about 3D-modeling of interior spaces 

Juho Kannala even directs a part of a new Academy of Finland project. Professor Simo Särkkä from the School of Electrical Engineering is responsible for the other part.

- The aim of the project is to do 3D-modeling in the interior spaces using a camera and inertial sensors. A robot or a human can prepare a map or a model. On the basis of the map it is possible to define a trajectory and a location of a device that has a camera, describes Kannala the new Academy of Finland project.

Kannala has graduated from the Helsinki University of Technology in 2004; in his Master’s Thesis project he measured and modeled sewer pipes by means of computer vision. Kannala attained his PhD degree in 2010 at the University of Oulu on the topic of  “Models and Methods for Geometric Computer Vision”.  After the doctoral defence he worked first as Academy of Finland Postdoctoral Researcher and then as Academy Research Fellow at the University of Oulu. Kannala came to the 911������ from the Center for Machine Vision Research at the Faculty of Information Technology and Electrical Engineering, University of Oulu.