Itziar Irigoien

Donostia International Physics Center, University of the Basque Country

Papers

Total Citations

H-Index

About

Itziar Irigoien is a researcher whose work bridges the fields of computer vision, robotics, and human-computer interaction, with a particular focus on gesture generation and activity recognition. Her most notable contribution is the development of a Generative Adversarial Network (GAN) framework for producing spontaneous talking gestures, a paper that has garnered 16 citations and addresses a critical challenge in creating more natural human-robot interactions. This work has implications for social robotics and autonomous navigation, where believable non-verbal communication is essential. Irigoien has also made contributions to video activity recognition, using Common Spatial Patterns to select relevant pixels for identifying human actions in surveillance footage—a task with significant real-world applications. Earlier in her career, she explored loop-closing in robotics through a typicality approach, demonstrating her sustained engagement with fundamental problems in autonomous systems. Her research consistently tackles the intersection of machine learning and embodied intelligence, aiming to make robots more perceptive and socially adept.