Andrea Torsello
Papers
2
Total Citations
33
H-Index
2
About
Andrea Torsello is a leading researcher in computer vision and 3D geometry processing, with a primary focus on the robust extraction of geometric primitives from unstructured point cloud data. His major contributions center on developing novel, clustering-based methodologies for cylinder detection, a critical challenge for applications in robotics, reverse engineering, and industrial inspection. Torsello’s work, including his highly cited 2020 paper on cylinder extraction as a clustering problem (23 citations), demonstrates a sophisticated approach that moves beyond traditional fitting techniques by leveraging pairwise axes similarities to achieve greater robustness in noisy, real-world environments. His 2019 paper (10 citations) further advances this framework, addressing the ubiquitous presence of cylindrical shapes in both natural and man-made scenes. By formulating cylinder estimation as a robust clustering problem, Torsello has provided a scalable and effective solution that significantly improves automated 3D analysis. His research is notable for its practical impact, offering a foundation for reliable geometric perception in autonomous systems and digital twin creation.
Research Focus
Key Achievements
Top Papers
- 1Cylinders extraction in non-oriented point clouds as a clustering problem23 citations · 2020
- 2Robust Cylinder Estimation in Point Clouds from Pairwise Axes Similarities10 citations · 2019