Imaging

In 1994, linguist Geoff Nunberg stated, in an article in the journal “Representations,” “reading what people have had to say about the future of knowledge in an electronic world, you sometimes have the picture of somebody holding all the books in the library by their spines and shaking them until the sentences fall out loose in space…” What would these fragments look like if you took every page of every book from 2.5 million volumes dating back over 500 years? Could every illustration, drawing, chart, map, photograph and image be extracted, indexed and displayed?

People can summarize a complex scene in a few words without thinking twice. It’s much more difficult for computers. But we’ve just gotten a bit closer -- we’ve developed a machine-learning system that can automatically produce captions (like the three above) to accurately describe images the first time it sees them. This kind of system could eventually help visually impaired people understand pictures, provide alternate text for images in parts of the world where mobile connections are slow, and make it easier for everyone to search on Google for images.

We present a model that generates free-form natural language descriptions of image regions. Our model leverages datasets of images and their sentence descriptions to learn about the inter-modal correspondences between text and visual data. Our approach is based on a novel combination of Convolutional Neural Networks over image regions, bidirectional Recurrent Neural Networks over sentences, and a structured objective that aligns the two modalities through a multimodal embedding.

In the first article of this series, I presented a timeline which clearly illustrates the increasing frequency of advocacy for librarians in DAM—a sentiment echoed across the board by practitioners, consultants, vendors, DAM news media, and academics.

IPTC announced today a new version of its Photo Metadata Standard, the most widely used standard to describe photos. It allows users to add precise and reliable data about people, products, locations and artwork shown in an image, and provides an improved and flexible way to express rights associated with a picture. IPTC is the world’s leading standards body for the news media and aims to simplify the distribution of information.

This paper presents an analysis of how humanities scholars utilize digital collections in their research and the ways in which digital collections could be enhanced for scholarly use. The authors surveyed and interviewed humanities faculty from twelve research universities about their research practices with digital collections and present analysis of the resulting responses.

The Smithsonian today unveiled the Smithsonian X 3D Collection and state-of-the-art 3-D explorer. The announcement kicked off the Smithsonian X 3D Conference, a two-day event focused on the current state of the Institution’s 3-D program and where it is headed in the future. A webcast of the conference is available.

At the Library of Congress, Dr. Lei He is an imaging scientist who is currently researching the effects of compression on digital images. Dr. He also uses quantitative methods to analyze “edges” found in images. “Edges” are naturally occurring high contrast areas of photographs that can be used to determine what resolution is needed for digitization. Dr. He’s research is already improving the processes at the Library of Congress.

The Library of Congress holds the world’s greatest collection of photographs. Mathew Brady’s Civil War classics, the portraits of American Indians by Edward S. Curtis and Carol Highsmith’s color views from the turn of the 21st century are just a few examples of the extraordinary photographs selected for inclusion in a new e-book, “Great Photographs from the Library of Congress.”

This SPEC Kit examines how research libraries and their parent institutions have responded to the transition from analog to digital images and the growth of digital images available from commercial vendors and/or created within institutions or their libraries. The survey gathers information about current practices relating to the development and management of institutional digital image collections and the acquisition and use of licensed image databases.