Mouna Selmi

Centre National de la Recherche Scientifique

Papers

Total Citations

H-Index

About

Mouna Selmi is a researcher in autonomous robotics and computer vision, with a focus on enabling humanoid robots to perceive and interpret human activities in real-world settings. Her most-cited work, "Vision-based Recognition of Activities by a Humanoid Robot" (2015, 9 citations), addresses a critical challenge: developing lightweight, robust activity recognition systems that function under the constraints of a moving robot platform with limited onboard computing. Unlike fixed-camera setups, Selmi’s approach accounts for the variability introduced by a robot’s own motion, making her contributions foundational for assistive robotics. By prioritizing computational efficiency without sacrificing accuracy, she has advanced the practical deployment of autonomous systems in human environments. Her research sits at the intersection of human-robot interaction, activity recognition, and resource-constrained machine learning. Though early in her career, Selmi’s work demonstrates a clear commitment to bridging the gap between theoretical computer vision and real-world robotic assistance—a crucial step toward robots that can safely and intelligently collaborate with people.