Visualizing Patterns in Databases of Cultural Images and Video

FAIN: HH-50002-09

Regents of the University of California, San Diego (La Jolla, CA 92093-0013)
Lev Manovich (Project Director: July 2008 to September 2011)

Consultation with the staff of the National Energy Research Scientific Computing Center to consider uses of computational analysis on large-scale photo and video collections.

To the best of our knowledge, our proposed project will be the first to combine theoretical perspectives of the humanities and cultural criticism with the analytical and visualization approach in order to study large datasets of images and video. Another innovation of our approach, which we call Cultural Analytics, is the use of image processing and computer vision techniques to analyze form and content of visual cultural data. We plan to process all images and video selected for our study to extract various features and mathematical descriptions of the visual structure. Consequently, we will use variety of statistical techniques including multivariate statistics methods such as Factor analysis, Cluster analysis, and multidimensional scaling to analyze this new metadata. Finally, we will use the results of our statistical analysis and the original data sets to produce a number of visualizations showing patterns in the data.