Marina Alberti
Papers
2
Total Citations
50
H-Index
2
About
Marina Alberti is a leading researcher in robotic perception and scene understanding, with a focus on integrating spatial reasoning with object recognition. Her key contributions lie in developing hybrid frameworks that combine top-down spatial relational reasoning with bottom-up object class recognition, significantly improving the reliability of perception systems in complex, real-world environments. Her most cited work (2014, 35 citations) demonstrates how leveraging scene context—such as the spatial relationships between objects—can enhance performance beyond what is achievable with intrinsic object features alone. In a subsequent study (2015, 15 citations), she systematically compared qualitative and metric spatial relation models, showing that incorporating diverse spatial cues into scene understanding pipelines reduces recognition errors. Alberti’s research addresses a critical limitation in robotics: the fragility of isolated object classifiers. By advancing context-aware perception, her work has implications for autonomous navigation, manipulation, and human-robot interaction. Her achievements include pioneering methods that bridge symbolic spatial reasoning with data-driven recognition, making her a notable figure in the intersection of computer vision and robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2