Logo des Repositoriums

Auflistung nach Schlagwort "emotion detection"

1 - 3 von 3
  • Konferenzbeitrag
    AudienceAR - Utilising Augmented Reality and Emotion Tracking to Address Fear of Speech
    (Mensch und Computer 2019 - Tagungsband, 2019) Hartl, Philipp; Fischer, Thomas; Hilzenthaler, Andreas; Kocur, Martin; Schmidt, Thomas
    With Augmented Reality (AR) we can enhance the reality by computer-generated information about real entities projected in the user's field of view. Hence, the user's perception of a real environment is altered by adding (or subtracting) information by means of digital augmentations. In this demo paper we present an application where we utilise AR technology to show visual information about the audience's mood in a scenario where the user is giving a presentation. In everyday life we have to talk to and in front of people as a fundamental aspect of human communication. However, this situation poses a major challenge for many people and may even go so far as to lead to fear and and avoidance behaviour. Based on findings in previous work about fear of speech, a major cause of anxiety is that we do not know how the audience judges us. To eliminate this feeling of uncertainty, we created an AR solution to support the speaker while giving a speech by tracking the audience's current mood and displaying this information in real time to the speaker's view: AudienceAR. By doing so we hypothesise to reduce the speaker's tension before and during presentation. Furthermore, we implemented a small web interface to analyse the presentation based on the audience mood after the speech is given. Effects will be tested in future work.
  • Konferenzbeitrag
    The Mood Game - How to use the player’s affective state in a shoot’em up avoiding frustration and boredom
    (Mensch und Computer 2019 - Tagungsband, 2019) Halbhuber, David; Fehle, Jakob; Kalus, Alexander; Seitz, Konstantin; Kocur, Martin; Schmidt, Thomas; Wolff, Christian
    In this demo paper, we present a shoot'em up game similar to Space Invaders called the "Mood Game" that incorporates players' affective state into the game mechanics in order to enhance the gaming experience and avoid undesired emotions like frustration and boredom. By tracking emotions through facial expressions combined with self-evaluation, keystrokes and performance measures, we have developed a game logic that adapts the playing difficulty based on the player's emotional state. The implemented game AI automatically adjusts the enemy spawn rate and enemy behavior, the amount of obstacles, the number and type of power ups and the game speed to provide a smooth game play for different player skills. The effects of our dynamic game balancing mechanism will be tested in future work.
  • Konferenzbeitrag
    SentiBooks: Enhancing Audiobooks via Affective Computing and Smart Light Bulbs
    (Mensch und Computer 2019 - Tagungsband, 2019) Ortloff, Anna-Marie; Güntner, Lydia; Windl, Maximiliane; Schmidt, Thomas; Kocur, Martin; Wolff, Christian
    We present SentiBooks, a smartphone application to enhance the audiobook listening experience via affective computing and smart light bulbs. Users can connect to Philips Hue Light Bulbs with a smartphone app while listening to an audiobook. The app analyzes the emotional expression of the narrator of the audiobook using speech emotion recognition and adjusts the colors of the lighting settings according to the expression of the narrator in 10-seconds intervals. By transitioning between colors that are connected to the specific emotion that is currently dominant in the reading, the overall audiobook experience is intensified.