PDF version of this entire document
Mauro explains that Diffusion geometries refer to the large-scale geometry of a manifold or a graph representing a data set, which is determined by long-time heat flows on the manifold/graph/data set. Other consider diffusion in another context.
From the abstract of [41]: In this paper, by incorporating spatially structured features into a histogram-based face-recognition framework, we intend to pursue consistent performance of face recognition. In our proposed approach, while diffusion distance is computed over a pair of human face images, the shape descriptions of these images are built using Gabor filters that consist of a number of scales and levels. It demonstrates that the use of perceptual features by Gabor filtering in combination with diffusion distance enables the system performance to be significantly improved, compared to several classical algorithms. The oriented Gabor filters lead to discriminative image representations that are then used to classify human faces in the database.
Gabor filter are also used by some leading algorithms for face recognition, so this approach might be worth exploring.
Roy Schestowitz 2012-01-08
