Dario Cioni

I am an ELLIS PhD student at Queen Mary University of London, supervised by Prof. Ioannis Patras and Prof. Georgios Tzimiropoulos, with Prof. Nicu Sebe (University of Trento) as ELLIS co-advisor, in collaboration with Tavus. My research focuses on deep generative models and multimodal learning for affective computing: building conversational agents that can perceive and reproduce the non-verbal richness of human interaction, such as facial expressions, gaze and voice intonation.

Previously, I was a Senior AI Engineer at the PwC Italy AI Center of Excellence, where I led the development of production-scale multimodal video generation systems and worked on multimodal perception and agentic AI systems. I hold an MSc in Artificial Intelligence from the University of Florence (2024) and have published at ICCV and ECCV workshops, earning a Best Workshop Paper Award at ECCV 2024.

I am always happy to discuss research and potential collaborations. Feel free to reach out by email.

news

May 01, 2026	I am beginning a PhD at Queen Mary University of London as part of the ELLIS PhD Program, supervised by Prof. Ioannis Patras and in collaboration with Tavus!
Sep 10, 2024	Our paper “Are CLIP features all you need for Universal Synthetic Image Origin Attribution?” received the Best Workshop Paper Award at the TWYN workshop at ECCV 2024!
Aug 10, 2024	One paper accepted at Trust What You learN (TWYN) workshop at ECCV 2024!
Aug 08, 2023	One paper accepted at 4th workshop on e-Heritage at ICCV 2023!

selected publications

ECCVW

Are CLIP features all you need for Universal Synthetic Image Origin Attribution?

Dario Cioni, Christos Tzelepis, Lorenzo Seidenari, and Ioannis Patras

In European Conference on Computer Vision Workshops (ECCVW), Trust What You learN (TWYN), 2024

Best Paper arXiv Bib PDF Code Website

Best Workshop Paper Award at the TWYN workshop, ECCV 2024.

@inproceedings{cioni2024clip,
  title = {Are CLIP features all you need for Universal Synthetic Image Origin Attribution?},
  author = {Cioni, Dario and Tzelepis, Christos and Seidenari, Lorenzo and Patras, Ioannis},
  booktitle = {European Conference on Computer Vision Workshops (ECCVW), Trust What You learN (TWYN)},
  year = {2024},
  eprint = {2408.09153},
  archiveprefix = {arXiv},
  primaryclass = {cs.CV},
}

ICCVW

Diffusion Based Augmentation for Captioning and Retrieval in Cultural Heritage

Dario Cioni, Lorenzo Berlincioni, Federico Becattini, and Alberto Del Bimbo

In IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2023

Bib HTML Code

@inproceedings{cioni2023diffusion,
  title = {Diffusion Based Augmentation for Captioning and Retrieval in Cultural Heritage},
  author = {Cioni, Dario and Berlincioni, Lorenzo and Becattini, Federico and Del Bimbo, Alberto},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)},
  year = {2023},
  pages = {1699-1708},
  doi = {10.1109/ICCVW60793.2023.00186},
}