Source Constrained Clustering - Robotics Institute Carnegie Mellon University

Source Constrained Clustering

Ekaterina Taralova, Fernando De la Torre Frade, and Martial Hebert
Conference Paper, Proceedings of (ICCV) International Conference on Computer Vision, pp. 1927 - 1934, November, 2011

Abstract

We consider the problem of quantizing data generated from disparate sources, e.g. subjects performing actions with different styles, movies with particular genre bias, various conditions in which images of objects are taken, etc. These are scenarios where unsupervised clustering produces inadequate codebooks because algorithms like K-means tend to cluster samples based on data biases (e.g. cluster subjects), rather than cluster similar samples across sources (e.g. cluster actions). We propose a new quantization technique, Source Constrained Clustering (SCC), which extends the K-means algorithm by enforcing clusters to group samples from multiple sources. We evaluate the method in the context of activity recognition from videos in an unconstrained environment. Experiments on several tasks and features show that using source information improves classification performance.

BibTeX

@conference{Taralova-2011-7404,
author = {Ekaterina Taralova and Fernando De la Torre Frade and Martial Hebert},
title = {Source Constrained Clustering},
booktitle = {Proceedings of (ICCV) International Conference on Computer Vision},
year = {2011},
month = {November},
pages = {1927 - 1934},
keywords = {clustering, quantization, k-means, bag-of-words, computer vision, activity recognition},
}