In this paper, we concentrate on the problem of finding the viewpoint that best satisfies a set of visual composition properties, often referred to as Virtual Camera or Viewpoint Composition. Previous approaches in the literature, which are based on general optimization solvers, are limited in their practical applicability because of unsuitable computation times and limited experimental analysis. To bring performances much closer to the needs of interactive applications, we introduce novel ways to define visual properties, evaluate their satisfaction, and initialize the search for optimal viewpoints, and test them in several problems under various time budgets, quantifying also, for the first time in the domain, the importance of tuning the parameters that control the behavior of the solving process. While our solver, as others in the literature, is based on Particle Swarm Optimization, our contributions could be applied to any stochastic search process that solves through many viewpoint evaluations, such as the genetic algorithms employed by other papers in the literature. The complete source code of our approach, together with the scenes and problems we have employed, can be downloaded from https://bitbucket.org/rranon/smart-viewpoint-computation-lib.

Improving the Efficiency of Viewpoint Composition

RANON, Roberto;
2014-01-01

Abstract

In this paper, we concentrate on the problem of finding the viewpoint that best satisfies a set of visual composition properties, often referred to as Virtual Camera or Viewpoint Composition. Previous approaches in the literature, which are based on general optimization solvers, are limited in their practical applicability because of unsuitable computation times and limited experimental analysis. To bring performances much closer to the needs of interactive applications, we introduce novel ways to define visual properties, evaluate their satisfaction, and initialize the search for optimal viewpoints, and test them in several problems under various time budgets, quantifying also, for the first time in the domain, the importance of tuning the parameters that control the behavior of the solving process. While our solver, as others in the literature, is based on Particle Swarm Optimization, our contributions could be applied to any stochastic search process that solves through many viewpoint evaluations, such as the genetic algorithms employed by other papers in the literature. The complete source code of our approach, together with the scenes and problems we have employed, can be downloaded from https://bitbucket.org/rranon/smart-viewpoint-computation-lib.
File in questo prodotto:
File Dimensione Formato  
Viewpoint_composition_IEEETVCG2014.pdf

non disponibili

Tipologia: Documento in Pre-print
Licenza: Non pubblico
Dimensione 2.83 MB
Formato Adobe PDF
2.83 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11390/1038563
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 24
  • ???jsp.display-item.citation.isi??? 14
social impact