Jan Puzicha, Joachim M. Buhmann, Yossi Rubner & Carlo Tomasi
Presented by: Dave Kauchak.
University of California, Department of Computer Science, San Diego. 47 slides.
The Problem: Image Dissimilarity
Where does this problem arise in computer vision?
Classification
Retrieval
Segmentation
Histograms for image dissimilarity
Histogram Example
Histogramming Image Features
Color
Texture
Gabor Filters
Creating Histograms from Features
Marginal Histograms
Cumulative Histogram
Dissimilarity Measure Using the Histograms
Heuristic Histogram Distances
More heuristic distances
Non-parametric Test Statistics
Cumulative Difference Example
Non-parametric Test Statistics (cont.)
Information-Theoretic diverges
Ground Distance Measure
Transportation Problem
Various properties of the metrics
Key Components for Good Comparison
Data Set: Color, Texture
Setup: Classification
Results
Conclusions