Événements du LITIS
- Cet évènement est passé
Séminaire « L’optimisation avec application en vision et segmentation d’images » par Ismail Ben Ayed
09/02/2016 @ 13:30 - 18:00
Pour la rentrée scientifique de l’équipe App, nous aurons l’honneur d’accueillir Ismail Ben Ayed (ETS Montréal) qui présentera ses travaux sur l’optimisation avec application en vision et segmentation d’images. Son séminaire devrait donc intéresser aussi nos collègues de STI et de QuantIF.
Le séminaire se tiendra le vendredi 2 septembre à 13h30 dans la bibliothèque du LITIS, 1er étage.
Venez nombreux à cet évènement de rentrée!
Title: High-Order Graphs in Computer Vision: A Pseudo-Bound Optimization Approach
Abstract: Recently, optimization of high-order and non-submodular graphical models has drawn a tremendous research interest in computer vision and machine learning. Such hard-to-optimize models arise naturally in a breadth of problems, e.g., data clustering, semantic segmentation, surface registration, classification, image deconvolution, high-order regularization, among many others. In this talk, I will discuss some recent developments in this direction, focusing on a general and powerful pseudo-bound optimization framework. I will discuss the key methodological aspects of this framework, and show how it improves the state-of-the-art. I will depict a rich array of problems and applications, including some graph clustering problems in computer vision as well as a variety of examples in the context of semantic segmentation (including several medical imaging examples).
Biography: Ismail Ben Ayed received the PhD degree (with the highest honor) in computer vision from the INRS-EMT, Montreal, QC, in 2007. He is currently Associate Professor at the École de Technologie Supérieure (ÉTS), where he holds an institutional research chair on artificial intelligence in medical imaging. Before joining the ÉTS, he worked for 8 years as a research scientist at GE Healthcare, London, ON, conducting research in medical image analysis. He also holds an adjunct professor appointment at Western University (since 2012). Ismail’s research interests are in computer vision, optimization, machine learning and their potential applications in medical image analysis. He co-authored a book and over seventy peer-reviewed publications, mostly published in the top venues in these subject areas. During his experience with GE, he received the GE innovation award (2010) and filed six US patents. Dr. Ben Ayed serves regularly as program committee member for the flagship conferences of the field, and as regular reviewer for the top journals.
Some publications related to the talk:
-M. Tang et al.: Normalized Cut meets MRFs, ECCV 2016 (to appear)
-I. Ben Ayed et al.: Distribution Matching with the Bhattacharyya Similarity: A Bound Optimization Framework. IEEE Trans. on Pattern Anal. Mach. Intell., 2015
-M. Tang et al.: Pseudo-bound Optimization for Binary Energies, ECCV 2014
-I. Ben Ayed et al.: Auxiliary Cuts for General Classes of Higher Order Functionals, CVPR 2013
-M. Tang et al.: Secrets of GrabCut and Kernel K-Means, ICCV 2015