General Framework for Subspace Detection
Welcome to the home page of the General Framework for Subspace Detection. Here, we present a general approach for detecting data alignments (subspaces) in large unordered noisy multidimensional datasets. In contrast to classical techniques such as the Hough transforms, which are designed for detecting a specific type of alignment on a given type of input, our approach is independent of the geometric properties of the alignments to be detected, as well as independent of the type of input data. Thus, it allows concurrent detection of multiple kinds of data alignments, in datasets containing multiple types of data.
The diagram presents the hierarchical representation of the generalization of detection techniques related to the Hough Transform. Our approach correspond to the green boxes (1) and (2). When applied to the detection of geometric shapes, our approches can be seen as the generalization of the Hough Transforms for analytic shapes (of arbitrary dimensionality) that can be represented by some linear subspace, e.g., (3-9). It is important to note that the techniques (10) and (11), i.e., the Generalized Hough Transforms, in yellow, target a different problem: the detection of shapes (in images) that cannot be represented analytically.
This project is being developed Leandro A. F. Fernandes and Manuel Menezes de Oliveira Neto.
Here you find all publicly available material related to this project.
