Grouping and Modeling of Repeated Imaged Scene Elements

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F15%3A00235481" target="_blank" >RIV/68407700:21230/15:00235481 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Grouping and Modeling of Repeated Imaged Scene Elements

Popis výsledku v původním jazyce

This proposal is to develop a robust method to group imaged repeated scene elements, infer semantic content from their groupings, e.g., complex repetitive patterns, and incorporate the rich structure and redundant information given by repetitive patternsinto scene models that are used for image retrieval, image geo- location and 3D reconstruction. A challenging testing dataset will be annotated and a loss function provided to assess the performance of new repeat grouping methods to compare against legacy approaches. Imaged repeated scene elements violate basic statistical independence assump- tions made by state-of-the-art approaches in image retrieval, geo-location, and 3D- reconstruction tasks. New models will be proposed that incorporate the repeated structure given by the proposed repeat grouping detection and repetitive pattern representation. The effect of explicitly accounting for repeated features will be demonstrated on standard datasets from the image retrieval and reconstru

Název v anglickém jazyce

Grouping and Modeling of Repeated Imaged Scene Elements

Popis výsledku anglicky

This proposal is to develop a robust method to group imaged repeated scene elements, infer semantic content from their groupings, e.g., complex repetitive patterns, and incorporate the rich structure and redundant information given by repetitive patternsinto scene models that are used for image retrieval, image geo- location and 3D reconstruction. A challenging testing dataset will be annotated and a loss function provided to assess the performance of new repeat grouping methods to compare against legacy approaches. Imaged repeated scene elements violate basic statistical independence assump- tions made by state-of-the-art approaches in image retrieval, geo-location, and 3D- reconstruction tasks. New models will be proposed that incorporate the repeated structure given by the proposed repeat grouping detection and repetitive pattern representation. The effect of explicitly accounting for repeated features will be demonstrated on standard datasets from the image retrieval and reconstru

Druh

V_souhrn - Souhrnná výzkumná zpráva

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

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)

Rok uplatnění

2015

Kód důvěrnosti údajů

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

Počet stran výsledku

34

Místo vydání

Praha

Název nakladatele resp. objednatele

Center for Machine Perception, K13133 FEE Czech Technical University

Verze

—