- machine learning
- deep learning
- computer vision
- data analytics
I joined what was then the Xeros Research Centre in July 2002 where I am currently a senior scientist, now at Naver Labs Europe, in the Computer Vision group.
My main research interests are image and video understanding, including document images, with a special focus on annotation, retrieval, semantic segmentation. I am also interested in the exploration of multi-modal information retrieval and fusion techniques as well as, more recently, in domain adaptation and transfer learning.
Short biography ( for more details see LinkedIn )
- Graduated from the Faculty of Mathematics and Computer Science of the University of Timisoara, Rumania (1991), I obtained the Master of Robotics and Vision (1992) and her Ph.D. degree (1996) in Computer Science (Science of engineering) from the University of Nice - Sophia Antipolis, France. My PhD thesis prepared within the group Robotvis, at INRIA, Sophia Antipolis studies the possibility of a projective representation of the three-dimensional environment using a pair of views obtained with uncalibrated cameras.
- I spent two years (1996 - 1998) as a postdoctoral associate within the groups Movi, INRIA Rhône-Alpes and Vista, IRISA Rennes where my research work focused on self-calibration and reconstruction from stereo image sequences and motion analyses of video sequences.
- Between 1998-1999 I was research and teaching assistant at CUI, University of Geneva, where within the Computer Vision Group, I was mainly involved in research on watermarking digital images and videos for copyright protection. I followed my research work on watermarking technologies as a research engineer in the Image and Video group of the Multimedia Communications Department, at Institut Eurécom (2000-2001).
- Contributed to around 100 scientific communications, including publications in top international conferences (CVPR, ICCV, ECCV, BMVC) and journals (PAMI, IJCV, CVIU, TOIS, MTA).
- Publications (ranked by citations) see Google Scholar.
- Most publications can be downloaded from ResearchGate.
- Main inventor of the bag-of-visual words (BOV) framework which revolutionized the field of computer vision and became a de facto standard for image classification for almost 10 years. The framework was first shared with partners in the LAVA project (IST-2001-34405) and published in the Deliverable 3.2 , then after being rejected from NIPS'03, it was presented at the SLCV workshop of ECCV'04.
- Associate Editor of EURASIP (Springer) Journal on Image and Video Processing (2011-2014) and Springer Signal, Image and Video Processing (since 2012).
- Peer reviewer for IEEE (PAMI, TIP, MM, TCSVT), Spinger (IJCV, MVA, SIVP, IR), Elsevier (PATREC, PR, IVC, JVLC, Imavis, Image, CMPB, SPIC), and others.
- Area Chair at ICCV 2011, Program Chair of IMAGAPP 2011, VISAPP 2012, KEPAF 2017.
- Main organizer of the ICPR LAVS'04 and VISAPP MMIU'08 Workshops.
- Occasionally, member in conference program committees (CVPR, ICCV, ECCV, ICIP, VISAPP) and PhD committees.
- Invited keynote speaker at VISAPP'10, KEPAF'11, ACIVS'15, and I gave several invited seminars.
- External referee of ERC Advanced Grant (2012) and ERC Starting Grant (2013) project proposals.
- External referee for National Research Agencies (ANR France, IST Israel, NWO Netherlands).
- Member of the Competitive selection jury of DR and CR2 at INRIA, France in 2015.
- Winner of ImageClef Domain Adaptation Challenge (2014).
- Winner of CLEF-IP Patent Image Classification and Image-based Patent Retrieval (2011).
- Winner of ImageClef Medical Image Modality Classification (2011, 2010) and Medical Ad-hoc Retrieval (2010).
- Winner of ImageClef Visual Concept Detection and Annotation Task (2010, 2009).
- Winner of ImageClef Wikipedia Photo Retrieval (2011, 2010).
- Winner of ImageClef Photo Retrieval (2009, 2008, 2007).
- Winner of the Pascal VOC Segmentation Challenge in 2008
- Second of the Pascal VOC Classification Challenge in 2007.