Mixture models and their applications

Content about mixture modelling

General purpose works:

Christophe Saint-Jean and Frank Nielsen. Online k-MLE for Mixture Modeling with Exponential Families.Proceedings of Geometric Science Information(GSI), LNCS 9389, pages 340-348, 2015.

Mixtures of Wishart distribution applied to motion clustering and recognition (2006- …):
```
              Example of clustering SPD matrices using Wishart model and Hartigan k-MLE
```
- Christophe Saint-Jean and Frank Nielsen. Hartigan’s method for k-MLE : Mixture modeling with Wishart distributions and its application to motion retrieval. “Geometric Theory of Information” pp 301-330 (Chap 11)., 2014. Christophe Saint-Jean and Frank Nielsen. A new implementation of k-MLE for mixture modelling of Wishart distributions. Proceedings of Geometric Science Information (GSI), LNCS 8085, pages 249-256, 2013. Sullivan Hidot and Christophe Saint-Jean. An expectation-maximization algorithm for the Wishart mixture model : Application to movement clustering. Pattern Recognition Letters, 31(14) :2318-2324, 2010. Sullivan Hidot, Christophe Saint-Jean, and Jean-Yves Lafaye. Classiﬁcation de signaux multidimensionnels utilisant la distribution de wishart : Application à la reconnaissance de mouvements. Actes de la conférence Groupement de Recherche en Traitement du Signal et des Images (GRETSI), pages 1321–1324, 2007. Sullivan Hidot, Jean-Yves Lafaye, and Christophe Saint-Jean. ACP relationnelle pour l’analyse du mouvement : Application à la danse. Actes de la conférence Reconnaissance des Formes et Intelligence Artiﬁcielle (RFIA), pages 64–72, 2006. Sullivan Hidot, Jean-Yves Lafaye, and Christophe Saint-Jean. Discriminant factor analysis for movement recognition : Application to dance. Machine Graphics and Vision, 15(3-4):391-399, 2006.

poster mixture of Wishart distributions for motion

Mixtures of von Mises distributions with application to reference-free measure of the quality of color images. Majed Chambah, Christophe Saint-Jean, and François Helt. Image quality evaluation in the ﬁeld of digital ﬁlm restoration. Proceedings of SPIE, Image Qualityand System Performance IV, volume 5668, pages 220–231, 2005.

Mixtures of gaussian components with robust estimation Christophe Saint-Jean and Carl Frélicot. A robust semi-supervised EM-based clustering algorithm with a reject option. Proceedings of the International Conference on Pattern Recognition (ICPR), volume 3, pages 399–402, 2002.