Optimizing multi-graph learning: towards a unified video annotation scheme
Proceedings of the 15th international conference on Multimedia
Image retrieval: Ideas, influences, and trends of the new age
ACM Computing Surveys (CSUR)
Leveraging user query log: toward improving image data clustering
CIVR '08 Proceedings of the 2008 international conference on Content-based image and video retrieval
Islamic geometrical patterns indexing and classification using discrete symmetry groups
Journal on Computing and Cultural Heritage (JOCCH)
Unified video annotation via multigraph learning
IEEE Transactions on Circuits and Systems for Video Technology
Supervised learning of similarity measures for content-based 3D model retrieval
LKR'08 Proceedings of the 3rd international conference on Large-scale knowledge resources: construction and application
Semantics extraction from images
Knowledge-driven multimedia information extraction and ontology evolution
Mutual information criteria for feature selection
SIMBAD'11 Proceedings of the First international conference on Similarity-based pattern recognition
Relevant gene selection using normalized cut clustering with maximal compression similarity measure
PAKDD'10 Proceedings of the 14th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining - Volume Part II
Localized graph-based feature selection for clustering
ICIAR'12 Proceedings of the 9th international conference on Image Analysis and Recognition - Volume Part I
Hypergraph based information-theoretic feature selection
Pattern Recognition Letters
Hi-index | 0.00 |
In this paper, we present a general guideline to establish the relation between a distribution model and its corresponding similarity estimation. A rich set of distance metrics, such as harmonic distance and geometric distance, is derived according to Maximum Likelihood theory. These metrics can provide a more accurate feature model than the conventional Euclidean distance (SSD) and Manhattan distance (SAD). Because the feature elements are from heterogeneous sources and may have different influence on similarity estimation, the assumption of single isotropic distribution model is often inappropriate. We propose a novel boosted distance metric that not only finds the best distance metric that fits the distribution of the underlying elements but also selects the most important feature elements with respect to similarity. We experiment with different distance metrics for similarity estimation and compute the accuracy of different methods in two applications: stereo matching and motion tracking in video sequences. The boosted distance metric is tested on fifteen benchmark data sets from the UCI repository and two image retrieval applications. In all the experiments, robust results are obtained based on the proposed methods.