Call for paper of a special issue of Image and Vision Computing
Title: Regularization Techniques for High-Dimensional Data Analysis
The explosive growth of high-dimensional visual data in computer vision requires effective techniques to reveal the underlying low-dimensional structure and discover the latent knowledge. Over the past decades, a variety of approaches for visual data modeling and analysis have been proposed, and representative methods include manifold learning, matrix factorization, subspace learning,sparse coding, and deep learning. However, these methods don’t perform well inmany practical applications when visual data contain various corruptions.Moreover, some of them are less theoretically interpretable. Recentdevelopments in regularization techniques have proven their effectiveness inhigh-dimensional visual data analysis, which are also with good interpretabilitiesin statistics. However, numerous problems in regularization techniques arestill unsolved and promising results may be achieved by incorporating theregularization techniques into the classical data modelling methods. Boththeoretical and technical developments are desirable to provide new insightsand tools in modelling the complexity of real world.
Thisspecial issue provides a forum for researchers all over the world to discusstheir works and recent advances in theory, algorithms, and applications forhigh-dimensional visual data analysis. we will invite one survey paper, whichwill undergo peer review. Papers addressing interesting real-world visualcomputing applications are especially encouraged. Topics relevant to thisspecial issue include, but are not limited to
l Modellinghigh-dimensional visual data with sparse, low rank, and other regularizedmethods
l Unsupervised,semi-supervised, and supervised regularized methods
l Dimensionreduction and subspace clustering with regularized methods
l Deepextensions of the existing regularized methods
l Largescale high-dimensional visual data modelling and analysis methods
l Robustrepresentation learning in high-dimensional data
l Multipledata modelling methods integration
l Real-worldapplications based on learning a compact model, e.g., face verification, metriclearning, object classification, scene recognition, multi-media dataabstraction, motion segmentation, etc.
l Novelmachine learning approaches for recovering the underlying low-dimensionalstructure from high-dimensional data, e.g., model selection, data completion,intrinsic dimensionality analysis, etc.
l Surveypapers regarding the topic of visual data analysis with regularization methods
l Paper submission: Jun. 15, 2016
l First notification: July 30, 2016
l Revision: Sep. 1,2016
l Final decision: Oct. 1, 2016
l Publication: Dec.15, 2016
Instructions for submission:
l To ensure that all manuscripts are correctly identified for inclusion into the special issue you are editing, it is important that authors select ‘SI:Regularization Techniques’ when they reach the “Article Type” step in the submission process. Please make sure authors are given this instruction when you send out invitation letters and/or instructions to potential authors.
Xi Peng, Institute for Infocomm., Research Agency for Science, Technology andResearch (A*STAR) Singapore, Email: firstname.lastname@example.org Weihong Deng, Beijing University of Posts and Telecommunications,China, Email: email@example.comAjmal Mian, School of Computer Science andSoftware Engineering, The University of Western Australia, Australia, Email: firstname.lastname@example.org