Authors: Haipeng Zeng, Xingbo Wang, Aoyu Wu, Yong Wang, Quan Li, Alex Endert, Huamin Qu
Abstract: Emotions play a key role in human communication and public presentations. Human emotions are usually expressed through multiple modalities. Therefore, exploring multimodal emotions and their coherence is of great value for understanding emotional expressions in presentations and improving presentation skills. However, manually checking and exploring the presentation videos is often tedious and time-consuming and there still lacks an effective tool to help users conduct an efficient and in-depth multi-level analysis. In this paper, we propose EmoCo, an interactive visual analytics system to facilitate efficient analysis of emotional coherence across text, facial and audio modalities in TED Talk videos. Our visualization system features a channel view and a projection view that together enable users to obtain a quick overview of emotional coherence and its temporal evolution. In addition, a detail view and word view enable detailed emotion exploration and comparison from the frame level and word level, respectively. We thoroughly evaluate the proposed system and visualization techniques through two usage scenarios and interviews with domain experts. The results demonstrate the effectiveness of our system in gaining insights into emotional coherence in presentations.