Multimedia Search Based on Non-Negative Matrix Factorization
スポンサーリンク
概要
- 論文の詳細を見る
A method is presented for similarity search based on dimensionality reduction by Non-negative Matrix Factorization (NMF). In similarity search based on NMF, the retrieval accuracy depends on the parameters e.g. the initial values of the factor matrices, the matrix rank and the number of iterations. In our method, we set the absolute value, of elements of the orthonormal matrices obtained by Singular Value Decomposition (SVD) to the initial values of the matrices in NMF. This initialization method helps us to determine the parameters in NMF. It is verified by experiments of image and video search that the retrieval accuracy by the proposed method is better than that by Latent Semantic Analysis (ISA).
- 社団法人電子情報通信学会の論文
- 2003-01-14
著者
-
URAHAMA Kiichi
Faculty of Design, Kyushu University
-
Hotta S
Department Of Computer And Information Sciences Nagasaki University
-
Urahama Kiichi
Faculty Of Design Kyushu University
-
Urahama K
Faculty Of Visual Communication Design Kyuslm Institute Of Design
-
Hotta Seiji
Department of Computer and Information Sciences, Nagasaki University
-
Urahama Kiichi
Faculty Of Computer Science And Systems Engineering Kyushu Institute Of Technology
関連論文
- バイラテラルフィルタを用いる再構成誤差最小化反復法による単一画像の超解像
- 非等方バイラテラルフィルタによる画像再構成を利用する単一画像の超解像
- 非等方バイラテラルフィルタによる画像再構成を利用する単一画像の超解像
- Graph-Spectral Filter for Removing Mixture of Gaussian and Random Impulsive Noise
- GENERATING ILLUSTRATION IMAGES WITH ISOLUMINANT COLORS(International Workshop on Advanced Image Technology 2006)
- IMAGE RECOLORING BY EIGENCOLOR MAPPING(International Workshop on Advanced Image Technology 2006)
- Multimedia Search Based on Non-Negative Matrix Factorization
- Robust Fuzzy Clustering of Similarity Data
- Graph-Spectral Filter for Removing Mixture of Gaussian and Random Impulsive Noise
- Denoising of Multi-Modal Images with PCA Self-Cross Bilateral Filter
- Semi-Supervised Classification with Spectral Subspace Projection of Data(Pattern Recognition)
- Unsupervised and Semi-Supervised Extraction of Clusters from Hypergraphs(Pattern Recognition)
- n-Mode Singular Vector Selection in Higher-Order Singular Value Decomposition
- Unsupervised Learning Algorithm for Fuzzy Clustering
- Semi-Supervised Classification with Spectral Projection of Multiplicatively Modulated Similarity Data(Pattern Recognition)
- Nonlinear Scale Spaces by Iterated Filtering of Images(Special Issue on Multiresolution Analysis)
- Equivalence between Fuzzy Hough Transform and Graph Spectral Method for Fuzzy Cluster Extraction
- Speed-Up of Image Analogy Method Based on Sparse Eigen-Decomposition
- Analog Method for Solving Combinatorial Optimization Problems
- Convergence of Alternative C-Means Clustering Algorithms
- Kth Largest Element Selection Circuit for Order Statistics Signal Processing
- Collaborative Filtering Based on Correlation between Items
- Edge-Preserving Cross-Sharpening of Multi-Modal Images
- Multimodal Pattern Classifiers with Feedback of Class Memberships
- Continuous Optimization for Item Selection in Collaborative Filtering(Biocybernetics, Neurocomputing)
- Acceleration of Image Retrieval Based on Edge Orientatioon Correlogram
- Non-iterative Symmetric Two-Dimensional Linear Discriminant Analysis