Alessio Tonioni

Google (Switzerland), University of Bologna

Papers

Total Citations

H-Index

About

Alessio Tonioni is a researcher at the forefront of robotic perception and 3D shape reconstruction, with a particular focus on integrating vision and touch. His key research areas span deep learning for robotics, tactile sensing, and semiautomatic data labeling. Tonioni’s most significant contribution is the development of TouchSDF, a novel DeepSDF approach that fuses vision-based tactile sensing with geometric deep learning to reconstruct 3D object shapes. This work, published in 2024, has already garnered 21 citations, highlighting its immediate impact on the field of robotic manipulation and haptics. By enabling robots to build comprehensive 3D models from tactile data, Tonioni is advancing how machines interact with and understand their physical environment. Earlier, he proposed the ARS (Augmented Reality Semiautomatic Labeling) method, which uses robotic camera movement and an augmented reality pen to efficiently generate large labeled datasets for deep learning in robotics. This work addresses a critical bottleneck in training robust perception models. Tonioni’s research bridges the gap between visual and tactile sensing, pushing the boundaries of autonomous robotic interaction.