Petia Radeva’s Homepage

Petia foto

Associate professor, at the Department MAIA, Faculty of MathematicsUniversitat de Barcelona, Icrea Academia

Senior Researcher at Computer Vision Center (CVC)

Head of Computer Vision at the University of Barcelona group, #1219, and  Barcelona Perceptual Computing Lab (BCNPCL)

Personal Web:,
email: petia dot ivanova at ub dot edu
Google Scholar CitationsMicrosoft Academic RecordLinkedIn, ResearchGate
A one-year postdoc position is open within an European project. For more details, see the link.
We have a vacant Master and/or PhD Fellowship on Lifelogging and Deep learning! If you are interested, send your CV and Academic transcript to


I did my undergraduate study at the University of Sofia, Bulgaria, at 1989. In 1991 I moved to Spain where in 1993 I presented my Master at the Universitat Autònoma de Barcelona in the field of Image Processing, Computer Graphics and Artificial Intelligence. In 1996, I received my Ph.D. degree from the Universitat Autònoma de Barcelona. Currently, I’m Head of Barcelona Perceptual Computing Laboratory (BCNPCL) at the University of Barcelona and Head of Medical Imaging Laboratory (MILab) of Computer Vision Center ( My present research interests are on development of learning-based approaches for computer vision and image processing. Some of the recent projects I’m currently leading or participate are: Can a monitorizing wearable camera help change our lifestyle?, Machine learning tools for large scale object recognition, Audience measurements by Computer Vision, Evaluation of Intestinal Motility by Endoluminal Image Analysis, Sponsored Research Agreement on Automatic Stent Detection in IVUS, Study for the development of a polyp detection algorithm under a Polyp Detection, etc.

Research interests

My research interests are in the fields of:

  • Lifelogging and egocentric vision
  • Event detection
  • Object discovery
  • Video segmentation and analysis
  • Image segmentation
  • Multi-class classification

Applications of Computer Vision and Machine Learning to Healthcare deserve special interest to me. In particular:

  • Applying life-logging to Mild Cognitive Impairment patients.
  • Applying life-logging to improve healthy habits of persons.
  • Wireless endoscopy for intestine motility disorders diagnosis.
  • Registration and retrieval of IVUS pullbacks for pre- post-intervention assessment.


Software license

Our API and APP for Automatic food recognition, LogMeal has been registered!

Selected Recent Publications

  • Lekadir K, Galimzianova A, Betriu A, Vila MD, Igual L, Rubin D, Fernandez

E, Radeva P, Napel S. A Convolutional Neural Network for Automatic Characterization of Plaque Composition in Carotid Ultrasound. IEEE J Biomed Health Inform. 2016 Nov 22. [Epub ahead of print]

  • Marc Bolaños, Álvaro Peris, Francisco Casacuberta, Petia Radeva, VIBIKNet: Visual Bidirectional Kernelized Network for Visual Question, submitted to IBPRIA’2016, arXiv:1612.03628 .

  • Alvaro Peris, Marc Bolaños, 3 , Petia Radeva, and Francisco Casacuberta: ICANN’16, Video Description using Bidirectional Recurrent Neural Networks. Proceedings of the 25th International Conference on Artificial Neural Networks (ICANN), 2016. pp. 3-11,>cs>arXiv: 1604.03390, 2016.








Screen Shot 2015-07-25 at 19.57.32Marc Bolaños, Ricard Mestre, Estefanía Talavera, Xavier Giró i Nieto, Petia RadevaVisual summary of egocentric photostreams by representative keyframes. ICME Workshops 2015: 1-6,>cs>arXiv:1505.01130, 2015.



  • Marc Bolaños, Mariella Dimiccoli, Petia Radeva, “TowardsStorytelling from Visual Lifelogging: An Overview”, arXiv:1507.06120, 2015, (under revision).




Screen Shot 2015-07-25 at 20.03.41


Marc Bolaños, Petia RadevaEgo-object discovery. CoRR abs/1504.01639, 2015.


Screen Shot 2015-07-25 at 20.08.18

Estefania Talavera, Mariella Dimiccoli, Marc Bolaños, Maedeh Aghaei and Petia Radeva. R-clustering for Egocentric Video Segmentation, IBPRIA’15, Santiago de Compostela, 2015.



Screen Shot 2015-07-25 at 20.05.44Marc Bolaños, Maite Garolera and Petia Radeva. Object Discovery using CNN Features in Egocentric Videos, IBPRIA’15, Santiago de Compostela, 2015.




motion-based segmentation

M. Bolaños, M. Garolera, P. Radeva, “Video segmentation of life-logging videos”, AMDO’2014, Springer-Verlag, 2014, p.1-9.




object discovery

M. Bolaños, P. Radeva, “Object Discovery for Egocentric Videos Based on Convolutional Neural Network Features”, Workshop on Storytelling, ECCV, 2014, (extended version submitted to a journal).



lifelogging tracking

M. Aghaei, P. Radeva, “Bag-of-Tracklets for Person Tracking in Life-LoggingData”, CCIA’2014, Proceedings of Catalonian Conference on Artificial Intelligence, Barcelona, Spain, 22 of October, 2014.




Maedeh Aghaei, Mariella Dimiccoli, Petia Radeva, Multi-Face Tracking by Extended Bag-of-Tracklets in Egocentric Videos, Computer vision and image understanding journal, 149 · July 2015, DOI: 10.1016/j.cviu.2016.02.013, arXiv:1507.04576

Active learning of dishesMarc Bolaños, Maite Garolera, Petia Radeva, “Active Labeling Application Applied to Food-Related Object Recognition”, 5th Workshop on Multimedia for Cooking and Eating Activities: CEA2013, ACM International Conference on Multimedia 2013, Barcelona October, 2013.


Ciompi F, Balocco S, Rigla J, Carrillo X, Mauri J, Radeva P. Computer-aided detection of intracoronary stent in intravascular ultrasound sequences. Med Phys. 2016 Oct;43(10):5616.

 Banchhor SK, Araki T, Londhe ND, Ikeda N, Radeva P, Elbaz A, Saba L, Nicolaides A, Shafique S, Laird JR, Suri JS. Five multiresolution-based calcium volume measurement techniques from coronary IVUS videos: A comparative approach. Comput Methods Programs Biomed. 2016 doi: 10.1016/j.cmpb.2016.07.009.

Araki T, Banchhor SK, Londhe ND, Ikeda N, Radeva P, Shukla D, Saba L, Balestrieri A, Nicolaides A, Shafique S, Laird JR, Suri JS. Reliable and Accurate Calcium Volume Measurement in Coronary Artery Using Intravascular Ultrasound Videos. J Med Syst. 2016 Mar;40(3):51. doi: 10.1007/s10916-015-0407-z. Oct;134:237-58.

Malagelada C, Drozdzal M, Seguí S, Mendez S, Vitrià J, Radeva P, Santos J, Accarino A, Malagelada JR, Azpiroz F. Classification of functional bowel disorders by objective physiological criteria based on endoluminal image analysis. Am J Physiol Gastrointest Liver Physiol. 2015 Sep 15;309(6):G413-9. doi: 10.1152/ajpgi.00193.2015.

Drozdzal M, Seguí S, Radeva P, Malagelada C, Azpiroz F, Vitrià J. Motility bar: A new tool for motility analysis of endoluminal videos. Comput Biol Med. 2015 Oct 1;65:320-30. doi: 10.1016/j.compbiomed.2015.04.006.

Araki T, Ikeda N, Dey N, Chakraborty S, Saba L, Kumar D, Godia EC, Jiang X, Gupta A, Radeva P, Laird JR, Nicolaides A, Suri JS. A comparative approach of four different image registration techniques for quantitative assessment of coronary artery calcium lesions using intravascular ultrasound Comput Methods Programs Biomed. 2015 Feb;118(2):158-72. doi: 10.1016/j.cmpb.2014.11.006.

Su-Lin LeeStefanie DemirciPetia RadevaGözde B. ÜnalEditorial note.Comp. Med. Imag. and Graph. 38(2): 69 (2014)

Francesco CiompiOriol PujolPetia RadevaECOC-DRF: Discriminative random fields based on error correcting output codes.Pattern Recognition 47(6): 2193-2204 (2014)

Michal DrozdzalSanti SeguíJordi VitriàCarolina MalageladaFernando AzpirozPetia RadevaAdaptable image cuts for motility inspection using WCE.Comp. Med. Imag. and Graph. 37(1): 72-80 (2013)

Francesco CiompiSimone BaloccoCarles CausJosepa MauriPetia RadevaStent Shape Estimation through a Comprehensive Interpretation of Intravascular Ultrasound Images.MICCAI (2) 2013: 345-352

Michal DrozdzalSanti SeguíPetia RadevaCarolina MalageladaFernando AzpirozJordi VitriàAn Application for Efficient Error-Free Labeling of Medical Images.Multimodal Interaction in Image and Video Applications 2013: 1-16

   Alberti MBalocco SCarrillo XMauri JRadeva P., “Automatic Non-rigid Temporal Alignment of Intravascular Ultrasound Sequences: Method and Quantitative Validation“, Ultrasound in Medicine and Biology, 39(9):1698-712.



Invited talk on “How Egocentric Vision and Visual Lifelogging complement medical images in health applications?!” at the SIPAIM’2016, Tandil, Argentina, 2016.

Invited talk on “What is common between Deep learning, Food Analysis and Egocentric Vision? at ICMV’2016, Nice, France, 18-30 of November, 2016.

Invited talk on “Can Deep Learning and Egocentric Vision for Visual Lifelogging help us eat better?” at the CCIA’16 conference, UPF, Barcelona, 20 of October, 2016.

Invited talk on “Lifelogging, egocentric vision and health: how a small wearable camera can help me improve my health state” 16 of September, 2016, Accenture Analytics, Sant Cugat del Vallés, Spain.

Invited talk on “Food Analysis by Deep Learning“, AMITANNS’2016, Albena, Bulgaria, June 22-27 2016.

Invited speaker at the ICMV’2015 Conference on: Deep learning and Lifelogging, 20 of November, 2015.

Invited speaker at the DIPP’2015 Conference: Digital Presentation and Preservation of Cultural and Scientific Heritage, talk on “Visual Lifelogging in the Era of Outstanding Digitization“, Veliko Tarnovo, Bulgaria, 28 of September, 2015.

Master thesis of Aniol Lidon on “Semantic and Diverse Summarization of Egocentric Photo Events“, 18 of September, 2015, Barcelona, Spain.

Talk at CAIP 2015:

CAIP 2015 Introduction on Visual Lifelogging I part

CAIP 2015 Deep learning II part

CAIP 2015 TemporalSegmentation III part

CAIP 2015 Social interactions IV part

Presentation at Bulgarian Academy of Science (BAN), “Object Discovery using CNN Features in Egocentric Videos”, Erasmus, August, 2015.

Talk on Visual Lifelogging: using non‐medical images for medical purposes at the Universitat Rovira i Virgili, Tarragona, Spain, 29 of Juliol, 2015.

Talk at CVPR’2015, Workshop on Medical imaging meets Computer Vision at the era of Big Data, Deep Learning and Novel Representations.

Talk “Life-logging and egocentric vision” at IMEC, Holst Center, Eindhoven, 2015, The Netherlands.

Talk  “Visual Lifelogging: using non-medical images for medical purposes”  during the PC meeting of MICCAI’2015, in Munich, May, 2015.

Talk at the University of Groningen2014 during my visit to the University of Groningen, The Netherlands, 24th of November, 2014.

Keynote speaker on “Talk Petia Radeva icmv2014“, of the 7th International Conference on Machine Vision, Milan, Italy, November 19-21, 2014.

Invited tutorial at the AMDO’2014 conference. Talk: “Lifelogging: what’s it about?“, 18 of July, 2014, Palma de Mallorca, Spain, 1.5 hours.

Invited talk organized within the European project “Advanced Computing for Innovation (Acommin). Topic: ”Video segmentation: applications to medical imaging and life-logging data”, January 14th 2014, Institute of information and communication technologies, bulgarian academy of sciences, Sofia, Bulgaria, 3 hours.

Invited talk organized within the European project “Advanced Computing for Innovation (Acommin). Topic: ”Advanced course on Computer Vision”, 24 to 26 July 2013, Institute of information and communication technologies, Bulgarian Academy of Science, Sofia, Bulgaria, 14 hours.


PhD and Master students:

Marc Bolaños (PhD Fellow, UB).

Maedeh Aghaei (PhD Fellow, APIF, UB) co-supervised with Dr. Mariella Dimiccoli (Postdoc, CVC).

Estefania Talavera (Master Fellow, UB, co-supervised with Prof. Nicolai Petkov, of the University of Groningen).

Mª Ángeles Jiménez (PhD Fellow, FPI, Ministry of Economy and Competence), co-supervised with Dr. Laura Igual (UB).

Mostafa Kamal, PhD student, co-supervised with Dr. Domenec Puig (URV), URV.

Marga Torre, La Generalitat de Catalunya, co-supervised with Dr. Fernando Hernandez (UPC) and Ricardo Toledo (UAB).

Eduardo Aguilar (PhD Fellow, Chile).

Gabriel de Oliveira (PhD Fellow).

Pedro Herruzo (Master student, UB).

Alejandro Cartas (PhD Fellow, Mexico).

Maria Leyva (Master student, UB).

PhD. Collaborators:

Mariella Dimiccoli (Postdoc, UB).

Beatriz Remeseiro, Juan de la Cierva Postdoc, UB.

Simone Balocco, Associate professor, UB.

Laura Igual, Associate professor, UB.


Visiting students:

Emanuela Molova, Master student, Bulgaria.

Yordan Petrov, Master student, Bulgaria.

External collaborations:

Maite Garolera, Head of Neuropsychology Unit at Consorci Sanitari de Terrassa.

Domenec Puig, Associate professor, Universitat Rovira i Virgili (URV).

Xavier Giro-i-Nieto, Associate professor, UPC.

Roberto Elosua, Senior researcher, IMIM.

Maria del Mar Vila, PhD student, IMIM.

Jose Massa & Jose Moreno, Universidad Nacional del Centro de la Provincia de Buenos Aires, Argentina.


Supervised PhD thesis:

15. Felip Miralles, 2016, UB

14. Michal Drozdzal, 2014, UB

13. Marina Alberti, 2013, UB

12. Francesco Ciompi, 2012, UB

11. Pierluigi Casale, 2011, UB

10. Sergio Escalera, 2009, UAB

9. David Rotger, 2008, UAB

8. Fernando Vilariño, 2006, UAB

7. Jaume Amores, 2006, UAB

6. Ignacio Pulido, 2005, University of Zaragoza

5. Misael Rosales, 2005, UAB

4. Debora Gil, 2004, UAB

3. Oriol Pujol, 2004, UAB

2. Cristina Cañero, 2003, UAB

1. Ricardo Toledo, 2002, UAB

Supervised Master thesis:

Chen Zang, BioHealth Master, co-supervised with Laura Igual (UB).

Marc Bolaños, UB, 2014.

Maedeh Aghaei, UB, 2013.


Currently, I’m teaching:

Computer Vision, a course for 3rd year students of Computer Science undergraduate study at the University of Barcelona.

Artificial Vision, a Master course for students of graduate study on Artificial Intelligence, UPC-UB-URV.

Advanced Medical Imaging, a Master course for students of graduate study on Biomedical Engineering.

Graduate course on Data science, UB.

Professional Service

Program committee of MICCAI’2015, CAIP’2015, CCIA’2015, etc.

Local area chair at CVPR’2014.

Local area chair of ICPR’2014.

Special sessions chairwomen of ICME’2014.

Chairwomen of Computing and Visualization for Intravascular Imaging and Computer Assisted Stenting (CVII-STENT), Workshop at MICCAI’2014.

Program committee member of: AMDO 2012CAIP 2009CAIP 2011CAIP 2013CAIP 2015CCIA 2007CCIA 2008CCIA’2014CCIA09CIARP 2012CIARP’2014CIARP2008CIARP2009HICV2011ICIAP2009IWITINCVPR 2011JCC2010, JCC2013, MCPR2010MCPR2013MCS2011MICCAI STENT 2013,  MICCAI-CVII 2011MICCAI-STENT’12,  MWPR2009,  RECPAT-2010.


Petia Radeva: «La transferència de coneixement és un motor de motivació per als investigadors», interview from the Fundació “Bosch i Gimpera”, May, 2016, Barcelona.

El CST rep una beca per la investigació en malalties neurodegeneratives“, terassadigital.cot, 31 d’Octubre, 2014.

“Convertir las 8 horas de película de la endoscopia sin cable en un ‘corto'”, El Mundo, April, 9th, 2014.


Algoritmos de visión artificial para la salud coronaria“, 08/01/2014,, Portal de ingenieros superiores.

“Algorithms of computer vision for coronary health”, UB, 2014.

UB premia estudis de neurolingüística i de nanotecnologia per vèncer càncer“, La Vanguardia, Desember 12, 2013.

La UB reclama més inversió pública per transferir el coneixement de les universitats“, EuropaPress, Desember 12, 2013.


Links of interest:

Quantified Self companies