< Back to list

SOM based artistic styles visualization

conferencePaper

DOI:10.1109/ICME.2013.6607474
Authors: Wang Ying / Takatsuka Masahiro

Extracted Abstract:

Painting collections from the old masters are valuable cultural heritage of human history. Their artistic styles can be generally determined by their art periods. From analyzing and visualizing the relationships of different artistic styles, information can be found to facilitate art history studies. In this paper, we propose a Self- organizing Map (SOM) based framework specifically for analyzing and visualizing the relationships among painting collections from artistic perspectives. In our framework, we first define a set of image features based on artistic concepts used in art criticism; then a SOM- based hierarchical model is used to analyze features extracted from individual artists’ painting collections. For our experiments, we obtain painting collections of six painting masters representing three art movements: post- impressionism, cubism and renaissance. An interactive web interface is also built to present our artistic influence analysis results. Through our experimental results, styles of different painting collections and their influential relationships can be analyzed and visualized from artistic perspectives. Index Terms—Artistic styles visualization, Artistic image features, Self-organizing map (SOM) 1.

Level 1: Include/Exclude

  • Papers must discuss situated information visualization* (by Willet et al.) in the application domain of CH.
    *A situated data representation is a data representation whose physical presentation is located close to the data’s physical referent(s).
    *A situated visualization is a situated data representation for which the presentation is purely visual – and is typically displayed on a screen.
  • Representation must include abstract data (e.g., metadata).
  • Papers focused solely on digital reconstruction without information visualization aspects are excluded.
  • Posters and workshop papers are excluded to focus on mature research contributions.
Show all meta-data