Object detection from multiple images based on the graph cuts

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F12%3A86084770" target="_blank" >RIV/61989100:27240/12:86084770 - isvavai.cz</a>

Výsledek na webu

<a href="http://download.springer.com/static/pdf/496/chp%253A10.1007%252F978-3-642-33179-4_26.pdf?auth66=1364565987_0df40d17670fc13265efea2784ac81a5&ext=.pdf" target="_blank" >http://download.springer.com/static/pdf/496/chp%253A10.1007%252F978-3-642-33179-4_26.pdf?auth66=1364565987_0df40d17670fc13265efea2784ac81a5&ext=.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-642-33179-4_26" target="_blank" >10.1007/978-3-642-33179-4_26</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Object detection from multiple images based on the graph cuts

Popis výsledku v původním jazyce

In this paper, we present a new method for detecting objects from multiple images. Unlike the general stereo-reconstruction methods, we propose a method for speci?c scenes only, which is motivated by the known fact that, in special circumstances, a specialized method will probably perform better than a general one. The theoretical model of the scene expects that the objects are created by the areas placed in a certain height over the scene base plane. The method is based on minimisation of an energy function. The function that has been found useful for the model is presented. For minimisation, the graph cut technique is used, which is modi?ed by an iterative process in which the original edge weights are changed on the basis of the residual graph. Themethod has been tested on detecting the cars in parking lots; examples of the results are presented.

Název v anglickém jazyce

Object detection from multiple images based on the graph cuts

Popis výsledku anglicky

In this paper, we present a new method for detecting objects from multiple images. Unlike the general stereo-reconstruction methods, we propose a method for speci?c scenes only, which is motivated by the known fact that, in special circumstances, a specialized method will probably perform better than a general one. The theoretical model of the scene expects that the objects are created by the areas placed in a certain height over the scene base plane. The method is based on minimisation of an energy function. The function that has been found useful for the model is presented. For minimisation, the graph cut technique is used, which is modi?ed by an iterative process in which the original edge weights are changed on the basis of the residual graph. Themethod has been tested on detecting the cars in parking lots; examples of the results are presented.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2012

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Lecture Notes in Computer Science. Volume 7431

ISBN

978-3-642-33178-7

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

10

Strana od-do

262-271

Název nakladatele

Springer

Místo vydání

Berlin

Místo konání akce

Rethymnon

Datum konání akce

16. 7. 2012

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—