Selecting local region descriptors with a genetic algorithm for real-world place recognition

Published in Conferences Papers
  1. Leonardo Trujillo, Gustavo Olague, Evelyne Lutton and Francisco Fernández de Vega. Multiobjective Design of Operators That Detect Points of Interest in Images. In Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation. 2008, 1299–1306. URL, DOI BibTeX

    	author = "Trujillo, Leonardo and Olague, Gustavo and Lutton, Evelyne and Fern\'{a}ndez de Vega, Francisco",
    	title = "Multiobjective Design of Operators That Detect Points of Interest in Images",
    	booktitle = "Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation",
    	series = "GECCO '08",
    	year = 2008,
    	isbn = "978-1-60558-130-9",
    	location = "Atlanta, GA, USA",
    	pages = "1299--1306",
    	numpages = 8,
    	url = "",
    	doi = "10.1145/1389095.1389344",
    	acmid = 1389344,
    	publisher = "ACM",
    	address = "New York, NY, USA",
    	keywords = "interest point detection, multiobjective optimization"

The basic problem for a mobile vision system is determining where it is located within the world. In this paper, a recognition system is presented that is capable of identifying known places such as rooms and corridors. The system relies on a bag of features approach using locally prominent image regions. Real-world locations are modeled using a mixture of Gaussians representation, thus allowing for a multimodal scene characterization. Local regions are represented by a set of 108 statistical descriptors computed from different modes of information. From this set the system needs to determine which subset of descriptors captures regularities between image regions of the same location, and also discriminates between regions of different places. A genetic algorithm is used to solve this selection task, using a fitness measure that promotes: 1) a high classification accuracy; 2) the selection of a minimal subset of descriptors; and 3) a high separation among place models. The approach is tested on two real world examples: a) using a sequence of still images with 4 different locations; and b) a sequence that contains 8 different locations. Results confirm the ability of the system to identify previously seen places in a real-world setting.

Published in
10th European Workshop on Evolutionary Computation in Image Analysis and Signal Processing (EvoASP'08)
Volume 4974
Pages 325-334
Date of conference
24 - 26 March 2008