ECCV2010 logo

Distance Functions and Metric Learning

Michael Werman Ofir Pele Brian Kulis

Abstract

Distance functions are at the core of numerous computer vision tasks. This tutorial provides an in-depth introduction to existing distances that can be used in computer vision and methods for learning new distances via metric learning. The tutorial has two parts. In the first part, we cover several existing distances, discuss their properties, and show how and when to apply them. We give special emphasis on Earth Mover's Distance variants and approximations, and efficient algorithms for their computation. In the second part, we cover metric learning methods; these methods aim to learn an appropriate distance/similarity function for a given task. We focus on learning Mahalanobis distance functions (a.k.a. quadratic forms), and also cover non-linear and non-Mahalanobis metric learning methods. Throughout this part, we will focus on a variety of existing vision applications of metric learning. For this tutorial, we only assume that attendees know some basic mathematics and computer science.


Where & When

ECCV 2010, Hersonissos, Crete                                                                              
Creta Maris Hotel, Minous south
Sunday 5 September 2010, 14:00-18:00

Material

* Part 1: Distance Functions
ppt
pdf
pdf(color 6 in pg)
pdf(b&w 6 in pg)
QC code
FastEMD code
* Part 2: Metric Learning
pdf