Andrea Caraffa

I'm a Research Scientist in Computer Vision and Deep Learning at Fondazione Bruno Kessler (FBK) currently working on foundation models and 6D pose estimation. I previously did research at SMaRC – Swedish Maritime Robotics Centre focusing on computer vision applications for autonomous underwater navigation.

My work focuses on developing novel deep learning approaches for challenging computer vision tasks. I'm particularly interested in zero-shot learning and fundamental deep learning research. Recently, I've been exploring how to leverage foundation models for 6D pose estimation without requiring task-specific training, with applications ranging from robotic manipulation to drone-based imaging.

I'm also passionate about creating datasets and benchmarks that push the boundaries of computer vision research. I've contributed to collecting and annotating datasets for wild berry detection using drones (WildBe) and thermal imaging of rural scenarios (MONET). My research aims to bridge the gap between theoretical advances in AI and real-world applications.