New Techniques for Visualizing Large Image and Video Collections

Lev Manovich — Sat, 31 Dec 2011 22:49:02 +0000

MW2012

Keywords:

visualization

Keywords:

images

Keywords:

video

Keywords:

analytics

Keywords:

digital image processing

Keywords:

UCSD

Abstract:

Many museums and media archives have already digitized large parts of their collections, and we may expect even more images and video to become digitally available in the near future. Are the standard web and desktop interfaces for accessing digital collections adequate for understanding these collections? Popular interfaces for showing images and video such as list, gallery, grid, and slide do now allow us to see the contents of a whole collection. These interfaces usually only display a few items at a time, regardless of whether you are in a browsing mode, or in a search mode. Because we are not able to see a collection as a whole, we can’t compare sets images or videos to each other, notice patterns of change over time, or understand parts of the collection in relation to the whole.

Graphing and visualization tools offer a range of visual techniques designed to reveal patterns in data. However, these tools have their own limitations. A key principle, which underlies the creation of graphs and information visualizations, is the representation of data using points, bars, lines, and similar graphical primitives. This principle has remained unchanged from the earliest statistical graphics of the early 19th century to contemporary interactive visualization software which can work with large data sets. Although such representations make clear the relationships in a data set, they also hide the objects behind the data from the user.

Since 2008, our lab (Software Studies Initiative, http://www.softwarestudies.com) at University of California, San Diego (UCSD) has been developing visual techniques that combine the strengths of media viewing applications and graphing and visualization applications. Like the latter, they create graphs to show relationships and patterns in a data set. However, while plot making software can only display data as points, lines or other graphic primitives, our software can show all the images in a collection superimposed on a graph. We call this method “media visualization.”

To date, we have already successfully applied our techniques to artworks, photos films, video games, comics, magazines, books, and other media content. Examples include all paintings by van Gogh, one million pages from manga series, hundreds of thousands of pages from 19th century American newspapers, and hundreds of hours of video gameplay. Our past and present collaborators include Getty Museum, Library of Congress, Austrian Film Museum, Magnum Photos, Netherlands Institute for Sound and Image and other institutions who are interested in using our methodology with their media collections. Starting in the Fall 2011, we are releasing our fully documented tools we developed under open source license.

In our demonstration, we will show the examples of our visualization and explain our key techniques for creating media visualizations.

Type:

Demonstration - show your project

MW2012 Demonstrations I

Site:

http://lab.softwarestudies.com/2008/09/cultural-analytics.html

Authors:

Lev Manovich

Rethinking Evaluation Metrics in Light of Flickr Commons

jdalton — Thu, 31 Mar 2011 15:15:07 +0000

MW2011

Paula Bray, and Sebastian Chan, Powerhouse Museum, Australi; Joseph Dalton, New York Public Library, USA; Dianne Dietrich, Cornell University Library, USA; Effie Kapsalis, Smithsonian Institution Archives, USA; Michelle Springer and Helena Zinkham, Library of Congress, USA

Abstract

In the past several years, cultural heritage institutions, including archives, libraries, and museums, have been placing their collections in Web spaces designed for collaboration and communication. Flickr Commons is one example of a highly visible space where cultural heritage institutions have partnered with a popular social networking site to provide greater discovery to, access of, and opportunities to interact with image collections on a large scale. It is important to understand how to measure the impact of these kinds of projects. Traditional metrics, including visit counts, tell only part of the story: much more nuanced information is often found in comments, notes, tags, and other information contributed by the user community. This paper will examine how several institutions on Flickr Commons - the Library of Congress, the Powerhouse Museum, the Smithsonian, New York Public Library, and Cornell University Library - are navigating the concept of evaluation in an emerging arena where compelling statistics are often qualitative, difficult to gather, and ever-changing.

Keywords: Flickr, metrics, evaluation, crowdsourcing, statistics, images

Program Item Reference:

Rethinking Evaluation Metrics in Light of Flickr Commons

museumsandtheweb.com - images

New Techniques for Visualizing Large Image and Video Collections

Rethinking Evaluation Metrics in Light of Flickr Commons

Paula Bray, and Sebastian Chan, Powerhouse Museum, Australi; Joseph Dalton, New York Public Library, USA; Dianne Dietrich, Cornell University Library, USA; Effie Kapsalis, Smithsonian Institution Archives, USA; Michelle Springer and Helena Zinkham, Library of Congress, USA

Abstract