Beschreibung:
Incorporating state-of-the-art techniques, this book provides complete theoretical and practical treatment of manifold learning. An excellent entry point for readers new to the subject, it supplies a high-level introductory view of the topic as well as in-depth discussion of the key technical details. It summarizes up-to-date advances in manifold learning. The comprehensive range of topics include basic theoretical background, implementation, and practical applications in fields such as medical, biometrics, multimedia, and computer vision.
Spectral Embedding Methods for Manifold Learning. Robust Laplacian Eigenmaps Using Global Information. Density Preserving Maps. Sample Complexity in Manifold Learning. Manifold Alignment. Large-scale Manifold Learning. Metric and Heat Kernel. Discrete Ricci Flow for Surface and 3-Manifold. 2D and 3D Objects Morphing Using Manifold Techniques. Learning Image Manifolds from Local Features. Human Motion Analysis Applications of Manifold Learning.