A Similarity Measure for Material Appearance

Autores/as

  • Manuel Lagunas Arto Graphics and Imaging lab, Universidad de Zaragoza
  • Sandra Malpica Graphics and Imaging lab, Universidad de Zaragoza
  • Ana Serrano Graphics and Imaging lab, Universidad de Zaragoza
  • Elena Garces Graphics and Imaging lab, Universidad de Zaragoza
  • Diego Gutierrez Graphics and Imaging lab, Universidad de Zaragoza
  • Belen Masia Graphics and Imaging lab, Universidad de Zaragoza

DOI:

https://doi.org/10.26754/jji-i3a.003500

Resumen

We present a model to measure the similarity in appearance between different materials, which correlates with human similarity judgments. We first create a database of 9,000 rendered images depicting objects with varying materials, shape and illumination. We then gather data on perceived similarity from crowdsourced experiments; our analysis of over 114,840 answers suggests that indeed a shared perception of appearance similarity exists. We feed this data to a deep learning architecture with a novel loss function, which learns a feature space for materials that correlates with such perceived appearance similarity. Our evaluation shows that our model outperforms existing metrics. Last, we demonstrate several applications enabled by our metric, including appearance-based search for material suggestions, database visualization, clustering and summarization, and gamut mapping.

Descargas

Publicado

2019-05-20

Número

Sección

Artículos (Tecnologías de la Información y las Comunicaciones)