Carlo Tomasi

Carlo Tomasi is the Iris Einheuser Distinguished Professor of Computer Science at Duke University's Trinity College of Arts & Sciences, with additional affiliations to the Duke Institute for Brain Sciences. His research integrates computer vision, machine learning, and applied mathematics, focusing on image motion analysis (NSF-funded), satellite image interpretation (IARPA-funded), computer-assisted diagnosis, and object recognition (Amazon-funded).

Research interests span optical flow estimation, video processing, autonomous systems, medical imaging, and biological applications. Recent work emphasizes transformer architectures, unsupervised learning for motion analysis, semantic integration in computer vision, and applications in both autonomous driving and ecological studies.

Publication trends show consistent focus on motion boundary detection (optical flow refinement, occlusion handling), multi-target tracking, and interdisciplinary applications in medical diagnostics and botany. Machine learning innovations include active learning strategies and noise-resistant network architectures.

Awards & Honors:

• ACM Fellow
• IEEE Computer Society Helmholtz Prize (awarded twice)

Significant grant support includes ongoing projects funded by NSF (image motion analysis), IARPA (satellite imagery), and Amazon (object recognition). Leads research groups focused on computer vision fundamentals and applied interdisciplinary collaborations with medical and biological researchers.