Oriol Barbany

/uRiɔl/ /baRbaɲ/

About me

I am a PhD student of the ELLIS PhD & PostDoc program at the Institut de Robòtica i Informàtica Industrial (IRI), where I am supervised by Prof. Carme Torras and Dr. Adrià Colomé. During 2024, I visited the École Polytechnique Fédérale de Lausanne (EPFL), where I was hosted by Prof. Amir Zamir .

Previously, I interned at Amazon, where I worked on computer vision and the learning-to-rank problem, and at Logitech, where I worked on voice conversion. I hold a Master's degree in Data Science from EPFL and a Bachelor's Degree in Telecommunications Technologies and Services Engineering from UPC.

Publications

OmniDiMM teaser figure showing 1D, 2D, and 3D modality outputs

OmniDiMM: Bridging 1D, 2D, and 3D with Any-to-Any Multimodal Modeling

Jason Toskov*, Oriol Barbany*, Rishubh Singh*, Jinya Sakurai, Efe Tarhan, Oğuzhan Fatih Kar, Roman Bachmann, Amir Zadeh, Jesse Allardice, Chuan Li, Carme Torras, Afshin Dehghan, Amir Zamir
arXiv preprint, 2026

Page BibTex

@article{omnidimm2026,
    title={{OmniDiMM: Bridging 1D, 2D, and 3D with Any-to-Any Multimodal Modeling}},
    author={Toskov, Jason and Barbany, Oriol and Singh, Rishubh and Sakurai, Jinya and Tarhan, Efe and Kar, O\u{g}uzhan Fatih and Bachmann, Roman and Zadeh, Amir and Allardice, Jesse and Li, Chuan and Torras, Carme and Dehghan, Afshin and Zamir, Amir},
    journal={arXiv preprint},
    year={2026},
}

BiFold robot bimanually folding a piece of cloth

BiFold: Bimanual Cloth Folding with Language Guidance

Oriol Barbany, Adrià Colomé, Carme Torras
IEEE International Conference on Robotics and Automation (ICRA), 2025

Page arXiv Code thread BibTex

@inproceedings{barbany2025bifold,
  author={Oriol Barbany and Adrià Colomé and Carme Torras},
  title={{BiFold: Bimanual Cloth Folding with Language Guidance}}, 
  year={2025},
  booktitle={2025 IEEE International Conference on Robotics and Automation (ICRA)}, 
  pages={5652-5659},
}

Multimodal search pipeline combining text and image queries

Leveraging Large Language Models for Multimodal Search

Oriol Barbany, Michael Huang, Xinliang Zhu, Arnab Dhua
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2024

Publication arXiv science thread BibTex

@Inproceedings{barbany24multimodal_search,
	author = {Oriol Barbany and Michael Huang and Xinliang Zhu and Arnab Dhua},
	title = {{Leveraging Large Language Models for Multimodal Search}},
	year = {2024},
	booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
}

ProcSim proxy-based confidence diagram for similarity learning

ProcSim: Proxy-based Confidence for Robust Similarity Learning

Oriol Barbany, Xiaofan Lin, Muhammet Bastan, Arnab Dhua
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024

Publication arXiv science thread BibTex

@InProceedings{barbany24procsim,
    author    = {Oriol Barbany and Xiaofan Lin and Muhammet Bastan and Arnab Dhua},
    title     = {{ProcSim: Proxy-Based Confidence for Robust Similarity Learning}},
    booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
    month     = {January},
    year      = {2024},
    pages     = {1308-1317}
}

Robot manipulating cloth in simulation next to its real-world counterpart

Benchmarking the Sim-to-Real Gap in Cloth Manipulation

David Blanco-Mulero, Oriol Barbany, Gokhan Alcan, Adrià Colomé, Carme Torras, Ville Kyrki
IEEE Robotics and Automation Letters, 2024

Page Publication Code arXiv thread BibTex

@article{blanco24benchmark,
  author={David Blanco-Mulero and Oriol Barbany and Gokhan Alcan and Adrià Colomé and Carme Torras and Ville Kyrki},
  journal={IEEE Robotics and Automation Letters},
  title={{Benchmarking the Sim-to-Real Gap in Cloth Manipulation}},
  year={2024},
  volume={9},
  number={3},
  pages={2981-2988},
}

Deformable surface reconstruction preserving Riemannian metric

Deformable Surface Reconstruction via Riemannian Metric Preservation

Oriol Barbany, Adrià Colomé, Carme Torras
Computer Vision and Image Understanding, 2024

Publication Code arXiv BibTex

@article{barbany24deformable,
  title = {Deformable surface reconstruction via Riemannian metric preservation},
  journal = {Computer Vision and Image Understanding},
  volume = {249},
  pages = {104155},
  year = {2024},
  issn = {1077-3142},
  doi = {https://doi.org/10.1016/j.cviu.2024.104155},
  url = {https://www.sciencedirect.com/science/article/pii/S1077314224002364},
  author = {Oriol Barbany and Adrià Colomé and Carme Torras},
}

Same ranked list rendered on mobile (top 3 visible) vs desktop (all visible), illustrating device-dependent position bias

Ranker-agnostic Contextual Position Bias Estimation

Oriol Barbany, Vito Bellini, Alexander Buchholz, Giuseppe Di Benedetto, Diego Marco Granziol, Matteo Ruffini , Yannik Steini
arXiv, 2021

arXiv BibTex

@misc{barbany21rankeragnostic,
	title={{Ranker-agnostic Contextual Position Bias Estimation}},
	author={Oriol Barbany Mayor and Vito Bellini and Alexander Buchholz and Giuseppe Di Benedetto and Diego Marco Granziol and Matteo Ruffini and Yannik Stein},
	year={2021},
	howpublished={arXiv:2107.13327},
}

Speaker A's voice converted to speaker B while the spoken content stays the same

FastVC: Fast Voice Conversion with non-parallel data

Oriol Barbany, Milos Cernak
Joint Workshop for the Blizzard Challenge and Voice Conversion Challenge, 2020

Page Publication arXiv BibTex

@inproceedings{barbany20fastvc,
	title = {{FastVC: Fast Voice Conversion with non-parallel data}},
	author = {Barbany, Oriol and Cernak, Milos},
	year = {2020},
	booktitle = {Joint Workshop for the Blizzard Challenge and Voice Conversion Challenge},
	pages = {145--149},
}

One vocoder model generating speech in three different speaker voices from the same text

Multi-Speaker Neural Vocoder

Oriol Barbany, Antonio Bonafonte, Santiago Pascual
IberSPEECH, 2018

Publication Code BibTex

@inproceedings{barbany18multispeaker,
	title = {{Multi-Speaker Neural Vocoder}},
	author = {Oriol Barbany and Antonio Bonafonte and Santiago Pascual},
	year = {2018},
	booktitle = {IberSPEECH},
	pages = {30--34},
}