Dissimilarity Detection of Two Video Sequences

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F13%3APU106367" target="_blank" >RIV/00216305:26230/13:PU106367 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.fit.vutbr.cz/research/pubs/all.php?id=10349" target="_blank" >http://www.fit.vutbr.cz/research/pubs/all.php?id=10349</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Dissimilarity Detection of Two Video Sequences

Popis výsledku v původním jazyce

This paper presents an approach for detection of differences between two visually identical video sequences. The video processing task for detection of short- and long-term changes between two video sequences is defined in detail. The algorithm comparing two video sequences (reference and query) is introduced together with definition of particular situations that the algorithm must be able to detect: re-written parts, removals or injected parts. The image processing methods are selected to be robust to several practical distortions that might appear in defined task. The appropriate computer-vision methods are presented and discussed, then proposed method and experiments are introduced and evaluated on manually generated dataset. Main focus of this work is on comparison of two different approaches for keyframe extraction: The first, more robust one is based on local features tracking, which we attempt to replace with computationally much less-expensive global descriptor approach with preservation of approximately the same video sequence dissimilarities detection success rate. Results of the different approaches are presented and discussed.

Název v anglickém jazyce

Dissimilarity Detection of Two Video Sequences

Popis výsledku anglicky

This paper presents an approach for detection of differences between two visually identical video sequences. The video processing task for detection of short- and long-term changes between two video sequences is defined in detail. The algorithm comparing two video sequences (reference and query) is introduced together with definition of particular situations that the algorithm must be able to detect: re-written parts, removals or injected parts. The image processing methods are selected to be robust to several practical distortions that might appear in defined task. The appropriate computer-vision methods are presented and discussed, then proposed method and experiments are introduced and evaluated on manually generated dataset. Main focus of this work is on comparison of two different approaches for keyframe extraction: The first, more robust one is based on local features tracking, which we attempt to replace with computationally much less-expensive global descriptor approach with preservation of approximately the same video sequence dissimilarities detection success rate. Results of the different approaches are presented and discussed.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2013

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of SCCG 2013

ISBN

978-80-223-3377-1

ISSN

1335-5694

e-ISSN

—

Počet stran výsledku

4

Strana od-do

28-31

Název nakladatele

Comenius University in Bratislava

Místo vydání

Smolenice

Místo konání akce

Smolenice

Datum konání akce

1. 5. 2013

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

—