A Neighborhood-Based Competitive Network for Video Segmentation and Object Detection

Authors:
Rafael Marcos Luque Baena;Enrique Dominguez;Domingo López-Rodríguez;Esteban J. Palomo
Affiliations:
Department of Computer Science, University of Málaga, Málaga, Spain;Department of Computer Science, University of Málaga, Málaga, Spain;Department of Applied Mathematics, University of Málaga, Málaga, Spain;Department of Computer Science, University of Málaga, Málaga, Spain
Venue:
ICANN '08 Proceedings of the 18th international conference on Artificial Neural Networks, Part I
Year:
2008

Citing 6
Cited 0

Pfinder: Real-Time Tracking of the Human Body

IEEE Transactions on Pattern Analysis and Machine Intelligence
Learning Patterns of Activity Using Real-Time Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
W4: Real-Time Surveillance of People and Their Activities

IEEE Transactions on Pattern Analysis and Machine Intelligence
Moving Shadow and Object Detection in Traffic Scenes

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 1
Learning and Removing Cast Shadows through a Multidistribution Approach

IEEE Transactions on Pattern Analysis and Machine Intelligence
Improved Production of Competitive Learning Rules with an Additional Term for Vector Quantization

ICANNGA '07 Proceedings of the 8th international conference on Adaptive and Natural Computing Algorithms, Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work proposes an unsupervised competitive neural network based on adaptive neighborhoods for video segmentation and object detection. The designed neural network is proposed to form a background model based on subtraction approach. The synaptic weights and the adaptive neighborhood of the neurons serve as a model of the background and are updated to reflect the statistics of the background. The segmentation performance of the proposed neural network is examined and compared to mixture of Gaussian models. The proposed algorithm is parallelized on a pixel level and designed to enable efficient hardware implementation to achieve real-time processing at great frame rates.