Massimiliano Mancini

Fondazione Bruno Kessler, Sapienza University of Rome, University of Tübingen

Papers

Total Citations

H-Index

About

Massimiliano Mancini is a researcher specializing in computer vision and machine learning, with a particular focus on robust visual recognition for robotics applications. His work addresses some of the most pressing challenges in deploying intelligent systems in real-world, unconstrained environments, including semantic place categorization, open world recognition, and domain adaptation. Mancini's most-cited contribution, "Learning Deep NBNN Representations for Robust Place Categorization" (2017, 34 citations), demonstrates how combining pretrained convolutional neural networks with part-based approaches can significantly advance visual place recognition. His influential 2019 work on web-aided open world recognition (26 citations) tackles a critical robotics limitation: the inability of trained systems to handle previously unseen objects, proposing innovative solutions that leverage web-sourced knowledge to bridge visual knowledge gaps. His research on domain adaptation, notably "Kitting in the Wild," further highlights his commitment to building vision systems that generalize across unpredictable working conditions. More recently, his work on open world recognition under shifting visual domains addresses the compounded challenge of detecting unknown concepts while navigating environmental variability. Collectively, Mancini's contributions push the boundaries of practical, adaptable visual intelligence for next-generation robotic systems.