Automatic Emotion Analysis in Movies: Matteo Garrone’s Dogman as a Case Study

This paper is the first step of an expansive ongoing initiative centered on automated film analysis through an ecocritical lens. Ecocriticism, an interdisciplinary field, delves into environmental themes within cultural works, broadening the scope of humanities’ focus on representation issues. Our objective is to pioneer a method for automated, dependable analysis of audiovisual narratives within fictional feature films, exploring the interplay between human emotions exhibited by characters and their surrounding environments. Using the acclaimed Italian crime/noir film, Dogman (2018), as a case study, we have constructed a modular pipeline integrating Facial Recognition and Emotion Detection technologies to scrutinize the emotional dynamics of the film’s two main characters. Our approach facilitates a comprehensive comparison over the film’s duration, enabling human analysts to future insights into the nuanced relationship between characters’ emotional states and the environmental contexts in which they unfold. Preliminary findings indicate promising outcomes from our pipeline, laying a solid foundation for subsequent film analyses. These results not only underscore the viability of automated methods in film studies but also offer a substantive starting point for deeper explorations into the complex interconnections between human emotions and cinematic environments.