Multimedia Knowledge and Social Media Analytics Laboratory

Chatzilari Elisavet

ehatzi@iti.gr

Dr. Elisavet Chatzilari received her diploma degree in Electronics and Computer Engineering from Aristotle university of Thessaloniki, Greece in 2008. She also holds a PhD degree on social media based scalable concept detection, University of Surrey (2014) (co-supervised by Prof. Josef Kittler (CVSSP – Surrey), Dr. Ioannis Kompatsiaris (ITI – CERTH) and Prof. Maria Petrou). She is currently a post-doctoral research fellow in Information Technologies Institute (ITI) at the Centre for Research & Technology Hellas (CERTH). She has participated in a number of EC-funded ICT projects (e.g. We-knowIT, Live+Gov, i-Treasures, MAMEM) and her research interests include image processing, multimodal analysis using social media, machine learning with large scale data (supervised learning, active learning, deep learning) and bio-signal processing. Her scientific work has been published in peer-reviewed journals, international conferences and book chapters.

 

Research:

http://mklab.iti.gr/research/salic

http://mklab.iti.gr/project/PerformancePrediction

Key Publications: 

Elisavet Chatzilari, Spiros Nikolopoulos, Yiannis Kompatsiaris, Josef Kittler, "SALIC: Social Active Learning for Image Classification", in IEEE Transactions on Multimedia, vol. 18, no. 8, pp. 1488-1503, Aug. 2016. doi: 10.1109/TMM.2016.2565440 [url] [code]

E. Chatzilari, S. Nikolopoulos, I. Patras, I. Kompatsiaris, "Leveraging social media for scalable object detection", Pattern Recognition, Volume 45, Issue 8, August 2012, Pages 2962-2979, DOI: 10.1016/j.patcog.2012.02.006.

A. Maronidis, E. Chatzilari, S. Nikolopoulos, I. Kompatsiaris, Scalable Image Annotation Using a Product Compressive Sampling Approach, IEEE DSAA 2015, Paris, 19-21 October 2015

Elisavet Chatzilari, Spiros Nikolopoulos, Yiannis Kompatsiaris, Josef Kittler, "How many more images do we need? Performance Prediction of bootstrapping for Image Classification", Proc. IEEE Int. Conf. on Image Processing (ICIP 2014), Paris, France, 27-30 October 2014

Elisavet Chatzilari, "Using Tagged Images of Low Visual Ambiguity to Boost the Learning Efficiency of Object Detectors", 21st ACM International Conference on Multimedia, Doctoral Symposium,  Barcelona, Spain, October 21-25, 2013. [prezi]

E. Chatzilari, S. Nikolopoulos, Y. Kompatsiaris, J. Kittler. "Multi-Modal Region Selection Approach for Training Object Detectors", In Proceedings of the 2nd ACM International Conference on Multimedia Retrieval (ICMR '12). ACM, New York, NY, USA, , Article 5 , 8 pages. DOI=10.1145/2324796.2324803.

Relevant Projects: