Real-time object detection using an evolutionary boosting strategy

被引:0
作者
Baro, Xavier [1 ]
Vitria, Jordi [1 ]
机构
[1] Ctr Visio Computador, Barcelona 08193, Catalonia, Spain
来源
ARTIFICIAL INTELLIGENCE RESEARCH AND DEVELOPMENT | 2006年 / 146卷
关键词
Boosting; genetic algorithms; dissociated dipoles; object detection; high dimensional feature space;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents a brief introduction to the nowadays most used object detection scheme. From this scheme, we highlight the two critical points of this scheme in terms of training time, and present a variant of this scheme that solves one of these points. Our proposal is to replace the WeakLearner in the Adaboost algorithm by a genetic algorithm. In addition, this approach allows Lis to work with high dimensional feature spaces which can not be used in the traditional scheme. In this paper we also use the dissociated dipoles, a generalized version of the Haarlike features used on the detection scheme. This type of features is an example of high dimensional feature space, moreover, when we extend it to color spaces.
引用
收藏
页码:9 / 18
页数:10
相关论文
共 8 条
  • [1] [Anonymous], 1996, ICML 96
  • [2] [Anonymous], INT J COMPU IN PRESS
  • [3] BALAS, 2003, J VISION, V3, P12
  • [4] BARO X, 2005, FRONTIERS ARTIFICIAL
  • [5] LIENHART R, 2002, P IEEE C IM PROC ICI, P155
  • [6] MCCANE B, 2003, IMAGE VISION COMPUT, P239
  • [7] Oza N. C., 2001, Online bagging and boosting, P229
  • [8] ULLMAN S, 1996, HIGH LEVEL VISION A