people

I am a final-year PhD Candidate in Artificial Intelligence at AImageLab, University of Modena and Reggio Emilia, advised by Prof. Roberto Vezzani.

My work sits at the intersection of Efficient Deep Learning and embedded systems: designing ultra-lightweight convolutional architectures and measuring what they really cost once deployed on NPUs, embedded GPUs and microcontrollers. Application domains range from aerial disaster response and surveillance to on-device fine-grained recognition and monocular depth estimation.

Before the PhD I obtained an M.Sc. in Artificial Intelligence Engineering cum laude (110/110) and spent six months as a research intern at Toyota Motor Europe in Belgium, working on transformer-based 3D human pose estimation, NeRF and Gaussian Splatting.

You can reach me at daniel.rossi@unimore.it, or browse my publications and projects.