Konstantina's Defense
Konstantina's PhD defense took place! Fun fact: I was supposed to be her opponent, but I'm happy I couldn't since we ended up being collaborators!
Recent Projects
Image Generation
Efficient and lightweight generation of images with desired characteristics by adapting the inference time behavior of Diffusion Models.
-
Merging and Splitting Diffusion Paths for Semantically Coherent Panoramas
Fabio Quattrini, Vittorio Pippi, Silvia Cascianelli, Rita Cucchiara
Alfie: Democratising RGBA Image Generation With No $$$Fabio Quattrini, Vittorio Pippi, Silvia Cascianelli, Rita Cucchiara
Handwriting Imitation
Algorithms for generating images of text with the desired content and handwriting style, methods to evaluate them, and application use cases.
-
Autoregressive Styled Handwritten Text Generation, but Make it Reliable
Carmine Zaccagnino Vittorio Pippi, Fabio Quattrini, Silvia Cascianelli, Alessio Tonioni, Rita Cucchiara
Zero-Shot Styled Handwritten Text Generation, but Make it AutoregressiveVittorio Pippi, Fabio Quattrini, Silvia Cascianelli, Alessio TOnioni, Rita Cucchiara
VATr++: Choose Your Words Wisely for Handwritten Text GenerationBram Vanherle, Vittorio Pippi, Silvia Cascianelli, Nick Michiels, Frank Von Reeth, Rita Cucchiara
Handwritten Text Generation from Visual ArchetypesVittorio Pippi, Silvia Cascianelli, Rita Cucchiara
HWD: A Novel Evaluation Score for Styled Handwritten Text GenerationVittorio Pippi, Fabio Quattrini, Silvia Cascianelli, Rita Cucchiara
How to Choose Pretrained HTR Models for Single Writer Fine-TuningVittorio Pippi, Silvia Cascianelli, Christopher Kermorvant, Rita Cucchiara
Document Understanding
Information extraction from 2D and 3D images of documents (from carbonized Roman papyri to modern ESGs), including ink bits, text, layout, and something about the writer.
-
μgat: Improving Single-Page Document Parsing by Providing Multi-Page Context
Carmine Zaccagnino, Fabio Quattrini, Silvia Cascianelli, Laura Righi, Rita Cucchiara
ToFu: Visual Tokens Reduction via Fusion for Multi-modal, Multi-patch, Multi-image TasksVittorio Pippi, Matthieu Guillaumin, Silvia Cascianelli, Rita Cucchiara, Maximilian Jaritz, Loris Bazzani
Volumetric FFC for Detecting Ink on the Carbonized Herculaneum PapyriFabio Quattrini, Vittorio Pippi, Silvia Cascianelli, Rita Cucchiara
Evaluating Synthetic Pre-Training for Handwriting Processing TasksVittorio Pippi, Silvia Cascianelli, Lorenzo Baraldi, Rita Cucchiara
The LAM Dataset: A Novel Benchmark for Line-Level Handwritten Text RecognitionSilvia Cascianelli, Vittorio Pippi, Martin Maarand, Marcella Cornia, Lorenzo Baraldi, Christopher Kermorvant, Rita Cucchiara
Boosting Modern and Historical HTR with Deformable ConvolutionsSilvia Cascianelli, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara
Mentorees
-
Vittorio Pippi
Vittorio has obtained PhD within the National PhD program in AI and he's just back from an internship at Amazon Research Berlin. I'm his proud co-supervisor. He's grounded, generous, and sharp. I could watch him learn to apply these traits to research and show them properly.
-
Fabio Quattrini
Fabio is a PhD student of the ICT program at the University of Modena and Reggio Emilia, now in internship at Amazon Research Berlin. I'm happy to be his advisor in this journey. He's creative, kind, and blunt. I'm seeing him gaining more and more control over his talent for research.
-
Carmine Zaccagnino
Carmine obtained his MSc degree from the University of Modena and Reggio Emilia. I've been his thesis co-tutor and we're still working together for his research internship at UniMoRe. He's hardworking, active, and passionate. I'm more and more pleased with his capacity to pick up on each opportunity he's presented with.
-
Kostantina Nikolaidou
Kostantina just obtained her PhD at the Luleå University of Technology. I've been collaborating with her since when I've been her advisor during her 3 month internship at UniMoRe. She's tenacious, resilient, and thoughtful. I'm amazed by her capacity to find and maintain connections, both professional and personal.
-
Pau Torras Coloma
Pau is pursuing his PhD at the Computer Vision Ceter of the Universitat Autònoma de Barcelona. I've been his advisor during his 1.5 month internship at UniMoRe we still collaborate. He's committed, curious, and fun. I'm delighted of his capacity to adapt to new working environments and approaches.
-
Bram Vanherle
Bram gained his PhD from the University of Hasselt and spent 5 months in internship at the University of Modena and Reggio Emilia, during which I had the pleasure of advising him. He's concrete, easygoing, and solid. I enjoyed a lot working with him (also after his internship).
-
Carlos Garrido Muñoz
Carlos is a PhD candidate at the University of Alicante. I've been his advisor during his 5 month internship at UniMoRe and we have been collaborating since then. He's proactive, independent, and resolute. I could observe how focused he can be when pursuing his goals.
External Links
Latest Updates
VisionDocs Workshop @WACV
I'm co-organizing the third edition of the Computer Vision Systems for Document Analysis and Recognition Workshop, this time at WACV2025! See you there!
AC/SPC @____
I'm serving as an Area Chair/Senior Program Chair for CVPR2026 and ECCV2026, AAAI2026 (first time!) and ICDAR2026 (first time!). Yay!
ELLIS Membership
I've officially become an ELLIS Member! It's amazing to be part of this spread but cohesive community!
Vittorio's Defense
Vittorio has been appointed the title of PhD! He did a great job and I know he'll do an even greater job in his next adventure!
CVPR2025
I've been so lucky to be at CVPR2025 in Nashville with almost my entire team. We definitely collected some fruits.
VisionDocs Workshop @ICCV
I'm co-organizing the second edition of the Computer Vision Systems for Document Analysis and Recognition Workshop, this time at ICCV2025! Get your papers ready!
AC @BMVC
I'm serving again as an Area Chair for BMVC2025! My soul is half British, so I hold this conference very dear!
Keynote @WACV-VisionDocs
I gave a [remote - sigh!] keynote at the VisionDocs workshop at WACV about how we teach machines to write and read: it's been an honor!
AI Governance Lab
I served as one of the lecturers for the AI Governance Lab course: it's always cool to teach some ML!
New InsAIde article out
Our ( Davide Morelli's and mine) Tech post of the InsAIde newsletter is out: have a look if you're curious!
ECCV2024
I'm just back from ECCV2024 in Milan. Some seeds have been planted: stay tuned to see how it goes!
AC @CVPR
It's official: I'll have the honor to serve as an Area Chair at CVPR2025!
AI4DH Workshop @ECCV2024
The workshop on AI for Digital Humanities I co-organize has reached its third edition at this ECCV2025!
Flash seminar @Huawei
I had the chance to chat with people at the Amsterdam Huawei Research Center about my research. I'm confident we gave each other interesting insights...
Bram's Defense
I had the pleasure of being a member of the jury for Bram Vanherle's PhD defense (and the chance to give a seminar at UHasselt the day before). Congrats to him!
ICDAR2024
I've concluded my summer with ICDAR2024 in Athens. That community feels like home each time!
ADAPDA Workshop @ICDAR2024
The Automatically Domain-Adapted and Personalized Document Analysis workshop I co-organize has had a very well-participated second edition at this ICDAR2024!
AC @ECCV
I'm serving as an Area Chair at ECCV2024! Get your papers ready!
Contact
Thanks for visiting my web portfolio! As you might have seen, I am an AI and Computer Vision Researcher and a people person and here's what (and where, and with whom) I am working on. If something sparked your interest or curiosity, please reach out!
If you see this face around, just wave and let's chatDrop me a message on LinkedIn
Send an e-mail to scascianelli[at]unimore[dot]it
Bye!