Silvia Cascianelli

Silvia Cascianelli is an AI and Computer Vision Researcher at the University of Modena and Reggio Emilia (UNIMORE). She actively contributes to the computer vision and document analysis communities through research, conference organization, and academic mentorship. She serves as Area Chair for major computer vision conferences including CVPR2025, BMVC2025, and ECCV2024, demonstrating her standing in the field.

Her research focuses on several key areas within computer vision and document analysis:

• Image Generation: Developing efficient and lightweight methods for image generation with desired characteristics, particularly using diffusion models
• Handwriting Imitation: Creating algorithms for generating images of text with specific content and handwriting styles, along with evaluation methods
• Document Understanding: Extracting information from 2D and 3D document images, ranging from modern documents to historical artifacts like carbonized Roman papyri

Dr. Cascianelli's work shows a clear progression toward more sophisticated generative models and evaluation frameworks, with recent publications focusing on diffusion models for handwritten text generation, efficient token reduction for multimodal tasks, and innovative approaches to historical document analysis. Her research bridges theoretical advancements with practical applications across diverse document types.

Her scientific contributions have been recognized through invitations to serve as Area Chair for top-tier computer vision conferences (CVPR, ECCV, BMVC) and opportunities to organize specialized workshops including VisionDocs at ICCV, AI4DH at ECCV, and ADAPDA at ICDAR.

• Area Chair at CVPR2025
• Area Chair at BMVC2025
• Area Chair at ECCV2024
• Organizer of VisionDocs Workshop at ICCV2025
• Organizer of AI for Digital Humanities Workshop at ECCV2024
• Organizer of ADAPDA Workshop at ICDAR2024

Dr. Cascianelli actively mentors the next generation of researchers:

• Vittorio Pippi - PhD Student at UniMoRe (National PhD program in AI)
• Fabio Quattrini - PhD Student at UniMoRe (ICT program)
• Carmine Zaccagnino - Research Intern at UniMoRe (formerly MSc student)
• Kostantina Nikolaidou - PhD Student at Luleå University of Technology
• Pau Torras Coloma - PhD Student at Computer Vision Center, Universitat Autònoma de Barcelona
• Bram Vanherle - CV Engineer at Colruyt Group Smart Innovation (formerly PhD student)

She is actively involved in several research initiatives including the AI Governance Lab where she serves as a lecturer, and collaborates with institutions worldwide. Her current projects focus on advancing diffusion models for image generation, improving handwritten text recognition systems, and developing novel methods for document understanding across historical and contemporary contexts.