Robust data whitening as an iteratively re-weighted least squares problem

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F17%3A00312642" target="_blank" >RIV/68407700:21230/17:00312642 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1007/978-3-319-59126-1_20" target="_blank" >http://dx.doi.org/10.1007/978-3-319-59126-1_20</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-319-59126-1_20" target="_blank" >10.1007/978-3-319-59126-1_20</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Robust data whitening as an iteratively re-weighted least squares problem

Popis výsledku v původním jazyce

The entries of high-dimensional measurements, such as image or feature descriptors, are often correlated, which leads to a bias in similarity estimation. To remove the correlation, a linear transformation, called whitening, is commonly used. In this work, we analyze robust estimation of the whitening transformation in the presence of outliers. Inspired by the Iteratively Re-weighted Least Squares approach, we iterate between centering and applying a transformation matrix, a process which is shown to converge to a solution that minimizes the sum of ℓ2 norms. The approach is developed for unsupervised scenarios, but further extend to supervised cases. We demonstrate the robustness of our method to outliers on synthetic 2D data and also show improvements compared to conventional whitening on real data for image retrieval with CNN-based representation. Finally, our robust estimation is not limited to data whitening, but can be used for robust patch rectification, e.g. with MSER features.

Název v anglickém jazyce

Robust data whitening as an iteratively re-weighted least squares problem

Popis výsledku anglicky

The entries of high-dimensional measurements, such as image or feature descriptors, are often correlated, which leads to a bias in similarity estimation. To remove the correlation, a linear transformation, called whitening, is commonly used. In this work, we analyze robust estimation of the whitening transformation in the presence of outliers. Inspired by the Iteratively Re-weighted Least Squares approach, we iterate between centering and applying a transformation matrix, a process which is shown to converge to a solution that minimizes the sum of ℓ2 norms. The approach is developed for unsupervised scenarios, but further extend to supervised cases. We demonstrate the robustness of our method to outliers on synthetic 2D data and also show improvements compared to conventional whitening on real data for image retrieval with CNN-based representation. Finally, our robust estimation is not limited to data whitening, but can be used for robust patch rectification, e.g. with MSER features.

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

<a href="/cs/project/LL1303" target="_blank" >LL1303: Vyhledávání vizuálních kategorií ve velkém množství obrázků</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2017

Kód důvěrnosti údajů

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

Název statě ve sborníku

Image Analysis

ISBN

978-3-319-59125-4

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

14

Strana od-do

234-247

Název nakladatele

Springer International Publishing

Místo vydání

Cham

Místo konání akce

Tromso

Datum konání akce

12. 6. 2017

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—