Marta Sanzari

Papers

Total Citations

H-Index

About

Marta Sanzari is a researcher at the intersection of robotics, computer vision, and human motion analysis. Her primary research areas include visual search and recognition for autonomous robot task execution, human motion primitive discovery, and the development of robust execution monitoring systems for assistive robots. In her most-cited work, "Deep Execution Monitor for Robot Assistive Tasks" (2019, 6 citations), Sanzari introduces a framework that leverages deep learning to monitor robot task performance in real-time, enabling safer and more adaptive human-robot collaboration. Complementing this, her paper "Visual search and recognition for robot task execution and monitoring" (2019, 6 citations) proposes a preliminary system that integrates visual search of environmental targets into the robot’s execution monitor, addressing a critical skill for autonomous agents. Earlier, in "Human motion primitive discovery and recognition" (2017, 4 citations), she presented a novel framework that automatically discovers and recognizes human motion primitives from motion capture data by optimizing a quantity called 'motion flux,' which depends on the motion of skeletal joints. This work contributes to the understanding of human movement patterns, with applications in rehabilitation, animation, and human-robot interaction. Sanzari’s research, though early in citation impact, demonstrates a clear trajectory toward enabling robots to perceive, interpret, and respond to human actions in assistive contexts.