Istituto di Scienza e Tecnologie dell'Informazione     
Salerno E., Wilson S., Bedini L., Kuruoglu E. E., Tonazzini A., Vatsa R. MUSCLE NoE - Computational learning methods for unsupervised segmentation. Multimedia Understanding through Semantics, Computation and Learning. Deliverable DN6.1, 2007.
We summarize the motivation and scope of our session on Computational Learning Methods for Unsupervised Segmentation (CLeMUS) at the KES 2007 Conference, and review the relationships between our Bayesian Component Separation view on unsupervised segmentation and recent accounts on soft segmentation based on statistical-probabilistic and/or variational approaches. Soft segmentation is more general than the classical partition-based approach to segmentation (hard segmentation), and its output can be reduced to a hard domain partition by applying some decision rule (e.g., thresholding or maximum likelihood). The papers presented at CLeMUS are reviewed and related to one another and to common concepts taken from recent literature.
Subject Machine Learning
Soft Segmentation
Blind Source Separation
I.4.6 Segmentation
I.4.8 Scene Analysis

Icona documento 1) Download Document PDF

Icona documento Open access Icona documento Restricted Icona documento Private


Per ulteriori informazioni, contattare: Librarian http://puma.isti.cnr.it

Valid HTML 4.0 Transitional