Using patch-based image synthesis to measure perceptual texture similarity

In: Computers & Graphics (2019), 81(104-116)
 

Abstract

The perceptual similarity of textures has gained considerable attention in the computer vision and graphics communities. Here, we focus on the challenging task of estimating the mutual perceptual similarity between two textures from materials on a consistent scale. Unlike previous studies that more or less directly queried pairwise similarity from human subjects, we propose an indirect approach that is inspired by the notion of just-noticeable differences (JND). Similar metrics are common in imaging and color science, but so far have not been directly transferred to textures, since they require the generation of intermediate stimuli. Using patch-based statistical texture synthesis, we produce continuous transitions between pairs of textures. In a user experiment, participants are then asked to locate an interpolated specimen in the linear continuum. Our intuition is that the JND, defined as the uncertainty with which participants perform this task, is closely related with the perceived pairwise texture similarity. Using a dataset of fabric textures, we show that this metric is particularly suitable to address fine-grained similarities, produces approximately interval-scale measurements and is additionally convenient for crowdsourcing. We further validate our approach by comparing it with a well-established data collection technique using the same dataset.

Keywords: Digital Material Appearance, Texture similarity, texture synthesis, Visual perception

Publication

This paper was published at the Journal of Computers & Graphics: https://doi.org/10.1016/j.cag.2019.04.001.

Images

Download Paper

Download Paper

Bibtex

@ARTICLE{Martin2019,
    author = {Mart{\'i}n, Rodrigo and Xue, Min and Klein, Reinhard and Hullin, Matthias B. and Weinmann, Michael},
     pages = {104--116},
     title = {Using patch-based image synthesis to measure perceptual texture similarity},
   journal = {Computers {\&} Graphics},
    volume = {81},
      year = {2019},
  keywords = {Digital Material Appearance, Texture similarity, texture synthesis, Visual perception},
  abstract = {The perceptual similarity of textures has gained considerable attention in the computer vision and
              graphics communities. Here, we focus on the challenging task of estimating the mutual perceptual
              similarity between two textures from materials on a consistent scale. Unlike previous studies that
              more or less directly queried pairwise similarity from human subjects, we propose an indirect
              approach that is inspired by the notion of just-noticeable differences (JND). Similar metrics are
              common in imaging and color science, but so far have not been directly transferred to textures,
              since they require the generation of intermediate stimuli. Using patch-based statistical texture
              synthesis, we produce continuous transitions between pairs of textures. In a user experiment,
              participants are then asked to locate an interpolated specimen in the linear continuum. Our
              intuition is that the JND, defined as the uncertainty with which participants perform this task, is
              closely related with the perceived pairwise texture similarity. Using a dataset of fabric textures,
              we show that this metric is particularly suitable to address fine-grained similarities, produces
              approximately interval-scale measurements and is additionally convenient for crowdsourcing. We
              further validate our approach by comparing it with a well-established data collection technique
              using the same dataset.},
      issn = {0097-8493},
       url = {http://www.sciencedirect.com/science/article/pii/S0097849319300421},
       doi = {https://doi.org/10.1016/j.cag.2019.04.001}
}