Noemi Canovi
Papers
1
Total Citations
2
H-Index
1
About
Noemi Canovi is a rising researcher at the intersection of computer vision, social robotics, and cognitive science, whose work focuses on how machines can interpret the nuanced, nonverbal cues that underpin human social interaction. Her key research area centers on the automated recognition of "vitality forms"—the subtle variations in action execution (e.g., gentle vs. abrupt movements) that communicate attitudes, intentions, and emotions. In her most-cited paper, "Diffusion-Based Unsupervised Pre-training for Automated Recognition of Vitality Forms" (2024), Canovi introduces a novel diffusion-based framework that learns to detect these expressive action signatures without requiring extensive labeled data. This unsupervised approach is a significant contribution, as it moves beyond simple action recognition (what is being done) to capturing how an action is performed—a critical step for building robots and AI systems that can engage in more natural, empathetic social exchanges. While her work is early-stage, with 2 citations to date, it represents a pioneering effort to bridge computational modeling with the psychological concept of vitality forms, originally developed by Daniel Stern. Canovi’s research promises to enhance human-robot interaction, making machines more attuned to the subtle dynamics of human communication.
Research Focus
Key Achievements
Top Papers
- 1