LS-N-IPS is an extension of the standard N-IPS particle filter (also known as CONDENSATION in the image processing literature). The modified algorithm adds local search to the baseline algorithm: in each time step the predictions are refined in a local search procedure that utilizes the most recent observation. A critical choice in the design of LS-N-IPS is the way the local search is implemented. Here, we introduce a method based on training artificial neural networks for implementing the local search. In experiments with real-life data (visual tracking) the method is shown to improve robustness and performance significantly, surpassing the …(more)
Please log in to take part in the discussion (add own reviews or comments).
Cite this publication
More citation styles
- please select -
%0 Conference Paper
%1 torma2003a
%A Torma, P.
%A Szepesvári, Cs.
%B 2003 IEEE International Symposium on Intelligent Signal Processing
%D 2003
%K application filtering, particle theory, vision,
%T Combining Local Search, Neural Networks and Particle Filters to Achieve Fast and Reliable Contour Tracking
%X LS-N-IPS is an extension of the standard N-IPS particle filter (also known as CONDENSATION in the image processing literature). The modified algorithm adds local search to the baseline algorithm: in each time step the predictions are refined in a local search procedure that utilizes the most recent observation. A critical choice in the design of LS-N-IPS is the way the local search is implemented. Here, we introduce a method based on training artificial neural networks for implementing the local search. In experiments with real-life data (visual tracking) the method is shown to improve robustness and performance significantly, surpassing the performance of previous state-of-the-art algorithms.
@inproceedings{torma2003a,
abstract = {LS-N-IPS is an extension of the standard N-IPS particle filter (also known as CONDENSATION in the image processing literature). The modified algorithm adds local search to the baseline algorithm: in each time step the predictions are refined in a local search procedure that utilizes the most recent observation. A critical choice in the design of LS-N-IPS is the way the local search is implemented. Here, we introduce a method based on training artificial neural networks for implementing the local search. In experiments with real-life data (visual tracking) the method is shown to improve robustness and performance significantly, surpassing the performance of previous state-of-the-art algorithms.},
added-at = {2020-03-17T03:03:01.000+0100},
author = {Torma, P. and Szepesv{\'a}ri, {Cs}.},
biburl = {https://www.bibsonomy.org/bibtex/272b1fb414bc545a98ddb220c1686f809/csaba},
booktitle = {2003 IEEE International Symposium on Intelligent Signal Processing},
date-modified = {2010-09-02 13:09:15 -0600},
interhash = {8335548fa72c0b3176876cec527dd5e1},
intrahash = {72b1fb414bc545a98ddb220c1686f809},
keywords = {application filtering, particle theory, vision,},
owner = {Beata},
pdf = {papers/lsnipsneuro.pdf},
timestamp = {2020-03-17T03:03:01.000+0100},
title = {Combining Local Search, Neural Networks and Particle Filters to Achieve Fast and Reliable Contour Tracking},
year = 2003
}