I. Introduction to Generalized Principal Component Analysis
(slides)
II. Basic GPCA Theory and Algorithms
(slides)
- Review of Principal Component Analysis (PCA)
- Introductory Cases: Line, Plane and Hyperplane Segmentation
- Segmentation with Known Number of Subspaces
- Segmentation with Unknown Number of Subspaces
III. Advanced Statistical Methods for GPCA
(slides)
- Lossy Coding of Samples from a Subspace
- Minimum Coding Length Principle for Data Segmentation
- Agglomerative Lossy Coding for Subspace Clustering
IV. Applications to Motion and Video Segmentation
(slides)
- 2-D and 3-D Motion Segmentation
- Temporal Video Segmentation
- Segmentation of Dynamic Textures
V. Applications to Image Representation and Segmentation
(slides)
- Multi-Scale Hybrid Linear Models for Sparse Image Representation
- Hybrid Linear Models for Image Segmentation
VI. Implementations
(link)