I received a PhD in Computer Science and Engineering from the University of Bologna in 2002 and currently I'm associate professor, with national scientific habilitation 09/H1 for the role of full professor, at the Department of Computer Science and Engineering, School of Engineering and Architecture, University of Bologna. My research interests include: computer vision, deep-learning, depth and scene flow estimation from monocular and stereo images, domain adaptation, and embedded computer vision [CV]

 

Short bio

Stefano Mattoccia

Associate Professor, PhD

Department of Computer Science and Engineering (DISI)
School of Engineering and Architecture
University of Bologna


Address: Viale del Risorgimento, 2  40136 Bologna, Italy

Phone:   +39 051 2093860  - Fax +29 051 2093869 

Email:     stefano.mattoccia#AT#unibo.it

Research

Keywords: computer vision, machine-learning, 3D vision, depth and scene flow estimation from monocular and stereo images, domain adaptation, embedded computer vision

My research activity is concerned with computer vision, machine learning applied to computer vision problems and embedded vision systems. These slides (2021) outline some of these research topics. For a more detailed overview take a look at the publication page or Google Scholar. My Orcid ID is: https://orcid.org/0000-0002-3681-7704


My team: Matteo Poggi (Assistant professor), Fabio Tosi (Postdoc researcher), Filippo Aleotti (PhD student), Youmin Zhang (PhD student), Fan Rizhao (PhD student), Huan Li (PhD student 2021), Andrea Conti (research fellow/PhD student 2021), Alessio Mingozzi (research fellow). Visiting PhD students: Xin Qiao (Xi'an Jiaotong University), Chaoqiang Zhao (East China University of Science and Technology, Shanghai)

   
   

 

Informativa sulla privacy disponibile a questo link

Workshop: Traditional Computer Vision in the Age of Deep Learning (TradiCV), ICCV 2021
Special issue:
Traditional Computer Vision in the Age of Deep Learning on Int. Journal of Computer Vision


Updates:

  1. -Best paper award honorable mention, 3DV 2021

  2. -Outstanding reviewer, BMVC 2021

  3. -Area Chair, CVPR 2022

  4. -1 paper accepted at ICCV 2021

  5. -1 paper accepted on IEEE Internet of Things Journal (2021)

  6. -3 papers accepted on IEEE Trans. on Pattern Analysis and Machine Intelligence (2021)

  7. -1 paper accepted on IEEE Trans. on Circuits and Systems for Video Technology (2021)

  8. -1 paper accepted at CVPR 2021

  9. -2 papers accepted at ECCV 2020

  10. -2 papers accepted at CVPR 2020


Recent selected papers:
- Neural disparity refinement for arbitrary resolution stereo, 3DV 2021

- Sensor-guided optical flow, ICCV 2021

- Learning optical flow from still images, CVPR 2021

- Self-adapting confidence estimation for stereo, ECCV 2020

- Reversing the cycle: self-supervised deep stereo through enhanced monocular distillation, ECCV 2020

- On the uncertainty of self-supervised monocular depth estimation, CVPR 2020

  1. -Distilled semantics for comprehensive scene understanding from videos, CVPR 2020

  2. -Real-Time Semantic Stereo Matching, ICRA 2020

- Learning end-to-end scene flow by distilling single tasks knowledge, AAAI 2020

Updates and recent results

  Youmin Zhang, University of Bologna/China Scholarship Council (2020-present)
      Depth sensing

  Fan Rizhao, University of Bologna /China Scholarship Council (2020-present)
      Depth sensing

  Filippo Aleotti, University of Bologna (2018-present)
      Unsupervised monocular depth estimation

  Fabio Tosi, University of Bologna (2017-present)
      Deep-learning for 3D reconstruction

      Matteo Poggi, University of Bologna (2014-2017)
      Machine learning for stereo vision

  Leonardo De-Maeztu Reinares, Public University of Navarre, Pamplona, Spain (2009-2012)
      Towards accurate and real-time local stereo correspondence algorithms: computational efficiency and massively
      parallel architectures
, Co-supervised with Arantxa Villanueva (Public University of Navarre)

  Federico Tombari, University of Bologna (2006-2009)
      Methodologies for visual correspondence, Co-supervised with Prof. Luigi Di Stefano (University of Bologna)

PhD students

Real-time single image depth perception in the wild with handheld devices: source code (iOS,Android) and web demo