Research on visual tracking is hampered due to the lack of standardised evaluation protocols and online reference repositories showing results on community accepted reference videos. Therefore, members of the project team co-organized IEEE Workshop on Visual Object Tracking Challenge in conjunction with International Conference on Computer Vision (ICCV 2013). The aim of the VOT2013 was to provide a common platform for comparison, analysis and discussion of existing and novel single object trackers. Organizers provided an evaluation kit that included videos with annotated ground truth as well as evaluation software to automate the experiments and calculate the performance measures.
B.01 Organiser of a scientific meeting
COBISS.SI-ID: 10409300Members of the project team have well established collaboration with sport researchers and experts. The collaboration consists mainly of transfer of methods and algorithms, developed in the field of computer vision into sport research and training. Result of this collaboration is a study of movement characteristics of ellite tennis players on a high-ranking ATP tennis tournament (2013 Valencia Open 500). In this instance, a computer vision based system for analysis of player motion, previously developed by several members of the project group was used for acquisition of player motion. It should be noted that such research is impossible to do without using computer vision methods.
F.17 Transfer of existing technologies, know-how, methods and procedures into practice
COBISS.SI-ID: 4355505Researchers, members of the project team have been awarded award for exceptional scientific achievements in 2012 for the paper: ČEHOVIN, Luka, KRISTAN, Matej, LEONARDIS, Aleš, Robust visual tracking using an adaptive coupled-layer visual model. IEEE Transactions on Pattern Analysis and Machine Intelligence DOI: 10.1109/TPAMI.2012.1
E.01 National awards
COBISS.SI-ID: 2926927