Emotion Recognition in Contemporary Art Installations

Modern art installations are becoming more and more use of computational systems to provide better interaction with the audience by reading the emotional reactions live. This paper outlines a detailed model of emotion recognition in immersive art setting that combines the theory of human feelings and the developed methods of artificial intelligence. Based on the basic models of emotions including the basic categories of Ekman, the wheel of Plutchik as well as multidimensional models of valence and arousal, the study finds a conceptual basis on which viewers internalize and articulate emotional states in their interactions with art. The suggested methodology will include the use of multimodal data collection of such practices as facial expression and voice tone, body movements and other physiological signs such as EEG. Those are fed to a hybrid deep learning pipeline that consists of Convolutional Neural Networks (CNNs) to extract visual attention and Long Short-Term Memory (LSTM) to extract temporal and physiological responses, making it possible to make the fine distinction between different emotions. Light modulation, spatial soundscapes, and projection mapping are integrated sensor technologies and interactive outputs in the implementation. An ongoing feedback system (AI) makes the installation more responsive to each individual viewer and turns the artwork into a live system that is capable of changing in response to the emotions of the audience

Affective Computing · Convolutional neural network · Emotion recognition · Emotional expression · Facial expression · Feeling · Aesthetic Perception and Analysis · Emotion and Mood Recognition · Face Recognition and Perception

• High-Resolution Image Synthesis with Latent Diffusion Models

  Open Access•Robin Rombach, Andreas Blattmann et al.•2022 IEEE/CVF Conference on…•2022

  Add to list