作者: Markus Mühling , Ralph Ewerth , Thilo Stadelmann , Bernd Freisleben , Rene Weber
关键词: Human–computer interaction 、 State (computer science) 、 Computer science 、 Game design 、 Computer game 、 Field (computer science) 、 Psychological research 、 Video content analysis 、 Game development tool 、 Multimedia 、 Graphical user interface
摘要: In this paper, we present an automatic semantic video analysis system to support interdisciplinary research efforts in the field of psychology and media science. The psychological question studied is whether how playing violent content computer games may induce aggression. To investigate question, extraction meaningful from required gain insights into interrelationship game events underlying neurophysiologic basis (brain activity) a player. Previously, human annotators had index according current state, which very time-consuming task. annotation large number recordings (i.e. videos) speeds up experimentation process allows researchers analyze more experimental data on objective basis. proposed for extracts several audiovisual low-level as well mid-level features deduces via machine learning approach. This requires manual annotations single only facilitate semi-supervised process. Finally, experts are allowed refine results graphical user interface. Experimental demonstrate feasibility