Marco San Biagio

Post Doc

Former colleague


Via Morego 30



I received my master degree cum laude in ”Informatic Engeneering for Intelligent Systems” from University of Palermo, Italy. Title of my thesis was: "Analysis of software and hardware redundancy techniques to improve reliability in microcontroller-based systems".

After that, I had a scholarship from GARR ( My work regarded the implementation and extension of services and network resources (VoIP and others services) of the CNR - National Research Council in Palermo.

In april 2014, I got my PhD in Nanoscienses (specifically in Computer Vision and Pattern Recognition) in collaboration with Università di Genova and the PAVIS department at the Italian Institute of Technology. The title of my ph.D. thesis was: "Statistical Data Fusion Techniques For Multi-cue Image Classification". This work regards the use of techniques that combine data from multiple sources (or sensors - e.g. thermal and visible image, audio and video) and gather this information in order to perform inferences, which will be more efficient and potentially more accurate than if they were achieved by means of a single source. 

Now I'm a Post-Doc at the PAVIS department at the Italian Institute of Technology, Genova. 

My research interests include computer vision, image/video processing, video-surveillance system, signal processing, statistical pattern recognition and data fusion techniques for object detection and classi cation.

Website 1:
Website 2:



Multi-modal Fusion of Video and Thermal Data, Object Detection, AVIOAero

Selected Publications


- San Biagio M. et al. «Recursive Segmentation Based on higher statistics in thermal imaging pedestrian detection», ISCCSP 2012 (oral presentation), IEEE.

- San Biagio M. et al. «Low-level multimodal integration on Riemannian Manifolds for Automatic Pedestrian Detection», Fusion 2012 (oral presentation), IEEE.

- San Biagio M. et al. «A Multiple Kernel Learning Approach to Multi-Modal Pedestrian Classification», ICPR 2012, IEEE.

- San Biagio M. et al. «Encoding Classes of Unaligned Objects Using Structural Similarity Cross-Covariance Tensors», CIARP 2013.

- San Biagio M. et al. «Heterogeneous Auto-Similarities of Characteristics (HASC): Exploiting Relational Information
for Classification», ICCV 2013, IEEE.

-San Biagio M. et al. «Weighted Bag Of Visual Words For Object Recognition», ICIP 2014, IEEE.

-San Biagio M. et al. «Encoding Structural Similarity By Cross-covariance Tensors For Image Classification», IJPRAI journal 2014, IEEE.