Main object of this work is to design and implement the algorithm for look direction determination, respectively the head movement. More specifically, it is a system that searches face in the video and then detects points, suitable for view direction estimation of tracked person. Estimation is realized using searching transformation, which has been performed on key points during head movement. For accuracy enhancement the calibration frames are used. Calibration frames determines the key points transformation in defined view directions. Main result is an application able to determine deflection of head from straight position in horizontal and vertical direction for tracked person. Output doesn't contain only information about deflection direction, but it also contains the size of deflection.

Unga barn föredrar redan från födseln att lyssna till positivt tal, vilket karaktäriserar barnriktat tal. Idenna studie kontrasterades neutral och positiv känsla i ett ordinlärningsparadigm där tittiden hos 16-månader gamla, svensktalande barn (N=12) analyserades när de testades på två påhittade ord.Målorden presenterades antingen i positivt barnriktat tal eller neutralt barnriktat tal. Detta gjordes föratt se om det positiva barnriktade talet underlättar inlärningen av nya ord. Deras tittid på detefterfrågade objektet analyserades sedan med en tvåvägs-ANOVA. Det fanns inga signifikantaskillnader mellan röstlägena. Slutsatsen som drogs var att hypotesen, att tittiden för det objekt sompresenteras i positivt tal är längre, inte kunde bekräftas men inte heller förkastas på grund av problemmed datainsamlingen samt dataanalysen. Inför framtida studier rekommenderas att utöka antaletdeltagare, samt undersöka andra variabler som kan påverka den tidiga ordinlärningen. / Already from birth young children prefer to listen to positive speech, which is a characteristic of childdirected speech. In this study neutral and positive tone was contrasted in a word-learning paradigmwhere the eye-movement of 16-month-old Swedish speaking children (N= 12) were analysed whenthey were tested on two made-up words. The words were either presented in neutral child-directedspeech or positive child-directed speech. This was done to see if the early word-learning is facilitatedby positive tone. The looking time was then analysed by a two-way ANOVA, and there was nosignificant difference between the two emotions. The conclusion drawn was that the hypothesis, thatthe looking time of the children that presented with positive child-directed speech would be longer,could not be confirmed nor rejected due to problems with collecting data and data analysis. Prior tofuture studies it is recommended to expand the number of participants, as well as examine othervariables that may affect early word learning.