Indexing Images for Visual Memory by Using {DNN} Descriptors -- Preliminary Experiments

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F14%3A00224957" target="_blank" >RIV/68407700:21230/14:00224957 - isvavai.cz</a>

Výsledek na webu

<a href="http://cmp.felk.cvut.cz/pub/cmp/articles/svoboda/Derner-TR-2014-25.pdf" target="_blank" >http://cmp.felk.cvut.cz/pub/cmp/articles/svoboda/Derner-TR-2014-25.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Indexing Images for Visual Memory by Using {DNN} Descriptors -- Preliminary Experiments

Popis výsledku v původním jazyce

Visual memory in mobile robotics is important to make the local- ization of a robot robust to situations, when GPS or similar localization methods are not available. Unlike many conventional approaches us- ing local features, we use a holistic method that employs deep neural networks (DNNs) to calculate a global descriptor of the whole image. We consider a scenario in which a robot equipped with an omni-directional camera calculates and stores DNN descriptors of images together with the positions as itmoves in the environment. When the position is unknown to the robot, the algorithm estimates it for a given omnidirectional image by matching it with the most similar database image. We compared our approach with a recently tested GIST-based approach onthe same dataset and we found out that the DNN-based approach yields better results. The experiments also show that the DNN-based algorithm is quite robust to partial occlusion, rotation and changes in lighting conditions.

Název v anglickém jazyce

Indexing Images for Visual Memory by Using {DNN} Descriptors -- Preliminary Experiments

Popis výsledku anglicky

Visual memory in mobile robotics is important to make the local- ization of a robot robust to situations, when GPS or similar localization methods are not available. Unlike many conventional approaches us- ing local features, we use a holistic method that employs deep neural networks (DNNs) to calculate a global descriptor of the whole image. We consider a scenario in which a robot equipped with an omni-directional camera calculates and stores DNN descriptors of images together with the positions as itmoves in the environment. When the position is unknown to the robot, the algorithm estimates it for a given omnidirectional image by matching it with the most similar database image. We compared our approach with a recently tested GIST-based approach onthe same dataset and we found out that the DNN-based approach yields better results. The experiments also show that the DNN-based algorithm is quite robust to partial occlusion, rotation and changes in lighting conditions.

Druh

V_souhrn - Souhrnná výzkumná zpráva

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2014

Kód důvěrnosti údajů

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

Počet stran výsledku

16

Místo vydání

Praha

Název nakladatele resp. objednatele

Center for Machine Perception, K13133 FEE, Czech Technical University

Verze

—