Published papers at CVPR, NeurIPS, and ICCV; two papers accepted at NeurIPS 2025, one on video similarity estimation and another on feature reliance of ImageNet-trained CNNs; a paper accepted at ICIAP 2025 on automatic benchmarking of large multimodal models; a paper accepted at ICCV 2025 on open-world image classification; a paper accepted at CVPR 2025 on person image synthesis.
Research Experience
Associate Professor at CIMeC/DISI - University of Trento; research projects include estimating video similarity, revisiting feature reliance of CNNs, automatic benchmarking of large multimodal models, and more.
Background
Associate Professor at the University of Trento, working in computer vision, machine learning, and multimodal AI. Research focuses on vision-language models and activity recognition, with applications in video analytics and industrial AI. Recently, exploring topics such as zero-shot action recognition, temporal action localization, and vocabulary-free image classification.
Miscellany
Co-founded Mountain Maps, a startup that uses AI to enhance outdoor navigation and help people explore mountain environments more safely and enjoyably.