Machine vision based facial expressions recognition and analysis for Filipino gamers

• Main Author: Sena, Juan Raphael
• Format: text
• Language: English
• Published: Animo Repository 2017
• Subjects: Video games > Philippines
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/5703
• Institution: De La Salle University

This research investigates the problem of expressions detection and analysis for Filipino gamers. The importance of expressions detection is that they are an advanced form of recognition that may be able to detect far more emotion than standard face tracking could do. These expressions would only last for a range of less than a second, which proves the difficulty of its analysis quite well. The aim of this Masters thesis is the development of a expression detection and analysis system that in order to understand user emotional response during an actual gameplay. Emotional response is one of the most important factors in video games, as it is a rare form of media that relies on interaction with the audience. The contribution to enhanced satisfaction in video games as well as improvement of interactive game play. A total of ten participants will be recruited in order to perform the experimentation and data gathering. Data will be collected using expression detection and analysis. Data analysis performed to generate results from questionnaires, psychological analysis, as well machine learning. This research is limited through the use of Open CV and Python for the coding, limited to the scope of the face for the detection, and limited to only detecting four levels of emotions. The contributions from the work include tools such as a new codebook or dictionary for learning a database of expressions, metrics for measuring performance, and systematic approach for detection.