Lorenzo Mur-Labadia

I am Lorenzo Mur-Labadia, a final-year PhD researcher in Deep Learning and Computer Vision at the University of Zaragoza (Spain), supervised by Prof. Rubén Martínez-Cantín and Prof. Josechu Guerrero.

My research focuses on video understanding, video–language learning, and 4D scene representations; with applications to embodied and egocentric perception.

During my PhD, I developed visual models for segmentation, detection, forecasting, and representation learning, leading to six first-author publications at top-tier computer vision venues, including ICRA 2023, ICCV 2023, IROS 2023, ECCV 2024, CVPR 2025, ICCV 2025 and 2 journals (CVIU and T-PAMI).

In the last 6 months, I was a Research Scientist intern at Meta AI (FAIR) in Paris, where I work with Adrien Bardes and Yann LeCun on large-scale self-supervised video understanding.

News

Jun 10, 2025	O-MaMa won the EgoVIS Object Correspondences Challenge at CVPR 2025.
Mar 11, 2025	DIV-FF: Dynamic Image-Video Feature Fields for Environment Understanding in Egocentric Videos was accepted at CVPR 2025 as a Highlight.
Jul 04, 2024	I participated in ICVSS 2024, held in Sicily.
Jul 01, 2024	AFF-ttention! Affordances and Attention Models for Short-Term Object Interaction Anticipation, developed at the University of Catania with A. Furnari and G. M. Farinella, was accepted at ECCV 2024.
Jun 17, 2024	I attended CVPR 2024 in Seattle, presenting two winning solutions from the Ego4D Challenge. Bayesian temporal-order priors for test-time refinement, in collaboration with C. Plou, won 1st Prize at the EgoVIS Step Grounding Challenge. AFF-ttention! Affordances and Attention Models for Short-Term Object Interaction Anticipation won 2nd Prize at the EgoVIS Short-Term Anticipation Challenge.
Nov 15, 2023	I was invited to the 3rd NAVER International Workshop on AI for Robotics.
Sep 06, 2023	I attended ICCV 2023 in Paris, presenting Multi-label Affordance Mapping from Egocentric Vision.

Publications

ICCV

O-MaMa: Learning object mask matching between egocentric and exocentric views

Lorenzo Mur-Labadia, Maria Santos-Villafranca, Jesus Bermudez-Cameo, and 3 more authors

In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025

Bib PDF

@inproceedings{mur2025mama,
  title = {O-MaMa: Learning object mask matching between egocentric and exocentric views},
  author = {Mur-Labadia, Lorenzo and Santos-Villafranca, Maria and Bermudez-Cameo, Jesus and Perez-Yus, Alejandro and Martinez-Cantin, Ruben and Guerrero, Jose J},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  pages = {6892--6903},
  year = {2025},
}

CVPR (HIGHLIGHT)

DIV-FF: Dynamic Image-Video Feature Fields For Environment Understanding in Egocentric Videos

Lorenzo Mur-Labadia, Josechu Guerrero, and Ruben Martinez-Cantin

In Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR HIGHTLIGHT), 2025

Bib PDF

@inproceedings{mur2025div,
  title = {DIV-FF: Dynamic Image-Video Feature Fields For Environment Understanding in Egocentric Videos},
  author = {Mur-Labadia, Lorenzo and Guerrero, Josechu and Martinez-Cantin, Ruben},
  booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR HIGHTLIGHT)},
  pages = {3470--3480},
  year = {2025},
}

Q2 JOURNAL

Temporal video segmentation with natural language using text-video cross attention and Bayesian order-priors

Carlos Plou, Lorenzo Mur-Labadia, Jose J Guerrero, and 2 more authors

Computer Vision and Image Understanding (CVIU), 2025

Bib

@article{plou2025temporal,
  title = {Temporal video segmentation with natural language using text-video cross attention and Bayesian order-priors},
  author = {Plou, Carlos and Mur-Labadia, Lorenzo and Guerrero, Jose J and Martinez-Cantin, Ruben and Murillo, Ana C},
  journal = {Computer Vision and Image Understanding (CVIU)},
  pages = {104622},
  year = {2025},
  publisher = {Elsevier}
}

Q2 JOURNAL

Random rotational embedding Bayesian optimization for human-in-the-loop personalized music generation

Miguel Marcos, Lorenzo Mur-Labadia, and Ruben Martinez-Cantin

PloS one Journal, 2025

Bib

@article{marcos2025random,
  title = {Random rotational embedding Bayesian optimization for human-in-the-loop personalized music generation},
  author = {Marcos, Miguel and Mur-Labadia, Lorenzo and Martinez-Cantin, Ruben},
  journal = {PloS one Journal},
  volume = {20},
  number = {11},
  pages = {e0335853},
  year = {2025},
  publisher = {Public Library of Science San Francisco, CA USA}
}

ECCV

Aff-ttention! affordances and attention models for short-term object interaction anticipation

Lorenzo Mur-Labadia, Ruben Martinez-Cantin, Jose J Guerrero, and 2 more authors

In European Conference on Computer Vision (ECCV), 2024

Bib PDF

@inproceedings{mur2024aff,
  title = {Aff-ttention! affordances and attention models for short-term object interaction anticipation},
  author = {Mur-Labadia, Lorenzo and Martinez-Cantin, Ruben and Guerrero, Jose J and Farinella, Giovanni Maria and Furnari, Antonino},
  booktitle = {European Conference on Computer Vision (ECCV)},
  pages = {167--184},
  year = {2024},
  organization = {Springer},
}

ICCV

Multi-label affordance mapping from egocentric vision

Lorenzo Mur-Labadia, Jose J Guerrero, and Ruben Martinez-Cantin

In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023

Bib PDF

@inproceedings{mur2023multi,
  title = {Multi-label affordance mapping from egocentric vision},
  author = {Mur-Labadia, Lorenzo and Guerrero, Jose J and Martinez-Cantin, Ruben},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  pages = {5238--5249},
  year = {2023},
}

IROS

Robust fusion for bayesian semantic mapping

David Morilla-Cabello, Lorenzo Mur-Labadia, Ruben Martinez-Cantin, and 1 more author

In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023

Bib PDF

@inproceedings{morilla2023robust,
  title = {Robust fusion for bayesian semantic mapping},
  author = {Morilla-Cabello, David and Mur-Labadia, Lorenzo and Martinez-Cantin, Ruben and Montijano, Eduardo},
  booktitle = {2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages = {76--81},
  year = {2023},
  organization = {IEEE},
}

ICRA

Bayesian deep learning for affordance segmentation in images

Lorenzo Mur-Labadia, Ruben Martinez-Cantin, and Jose J Guerrero

In 2023 IEEE International Conference on Robotics and Automation (ICRA), 2023

Bib PDF

@inproceedings{mur2023bayesian,
  title = {Bayesian deep learning for affordance segmentation in images},
  author = {Mur-Labadia, Lorenzo and Martinez-Cantin, Ruben and Guerrero, Jose J},
  booktitle = {2023 IEEE International Conference on Robotics and Automation (ICRA)},
  pages = {6981--6987},
  year = {2023},
  organization = {IEEE},
}