Hi, this is Marco 👋 I love to learn and dream big (after all, it costs zero!), and I’m always looking for new challenges and opportunities to grow.
I write here to share my thoughts, ideas, and projects.
Mostly related to technology and artificial intelligence.
🎓 Education and working experience
I work at EVS Embedded Vision Systems, Srl (Verona, Italy) as Senior Research Scientist since 2019, collaborating with prof. Vittorio Murino.
I’ve received my PhD in Computer Science in 2020 within the Vision, Image Processing & Sound (VIPS) Lab at the University of Verona, Italy, under the supervision of prof. Marco Cristani.
During my studies, I’ve been working on Machine Learning (Deep Learning) focusing on the exploitation of visualization techniques to understand the internal representation of neural networks, with particular attention to CNNs for image classification and regression. My research aims to identify and use relevant information from 2D images and 3D volumes, with active recognition systems in mind.
From July 2018 to March 2019, I collaborated with the Istituto Italiano di Tecnologia, Genova, Italy, prof. Alessio Del Bue. I also spent three years (2012-2015) as research assistant in the ALTAIR Robotics Lab (Verona), prof. Paolo Fiorini, where I was involved in the SAFROS and I-SUR European Projects (FP7-ICT), working on the development of a surgical simulator and a tracking system of anatomical structures.
My background also includes Computer Vision and graphics applied to surgical robotics, augmented reality and game development.
📜 Publications
- Saliency-based approaches for multidimensional explainability of deep networks
2020, PhD thesis - Active 3D Classification of Multiple Objects in Cluttered Scenes
2019, ICCV Workshop - 6D Pose Estimation for Industrial Applications
2019, ICIAP (Industrial Session) - Estimating body fat from depth images: hand-crafted features vs Convolutional Neural Networks
2018, 3D BODY.TECH - Analyzing Body Fat from Depth Images
2018, 3DV - Recognition self-awareness for active object recognition on depth images
2018, BMVC - Understanding Deep Architectures by Visual Summaries
2018, BMVC - A Robust Particle Filtering Approach with Spatially-Dependent Template Selection for Medical Ultrasound Tracking Applications
2016, VISAPP - Robust 3D Pose Estimation of a Laparoscopic Instrument with three Landmarks
2015, STAG (Eurographics Italian Chapter)