This repository holds an archive of the materials produced for the ADDI project, which ran from May 2021 through January 2022. The project was designed to adapt and apply computer vision algorithms to aid in the discovery and use of digital collections, specifically documentary photography collections held by the Library of Congress. The five collections that were used in this study are:
All of the collections have been digitized by the Library of Congress and are listed as having “No known restrictions on publication.”
Each of the final elements of the project are listed in this repository in one of the number folders. These are (1) the replication code, (2) the computed metadata, (3) a methods paper, (4) a data analysis paper, (5) biweekly reports submitted to the Library of Congress, (6) code for an interactive visualization of the metadata, and (7) an archived version of the Distant Viewing Toolkit.
A working version of the visualization tool can be found through GitHub pages at:
Versions of the two papers can be viewed online here:
The methods and data analysis papers are released under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0). All of the code is released under the GNU GPL v2.0.
This project was funded by the Library of Congress through a Broad Agency Announcement (LCCIO20D0112) in support of Computing Cultural Heritage in the Cloud, a project funded by the Andrew W. Mellon foundation.
For questions, please contact the directors of the Distant Viewing Lab: Taylor Arnold (tarnold2@richmond.edu) and Lauren Tilton (ltilton@richmond.edu). Note that this repository is for archival purposes and will not be regularly updated. For active development on the Distant Viewing Toolkit see its dedicated GitHub repository.
We are grateful to the Library of Congress and LC Labs for the opportunity to be a part of this innovative initiative. Thank you to the teams across LoC that met with us and graciously shared their knowledge and expertise. In particular, the Prints & Photographs Division team took significant time to guide us through their collections and histories of digitization, which was invaluable knowledge that shaped ADDI. This project is only possible because of the library’s pioneering work to collect, preserve, and digitize photographs. Thank you as well to Meghan Ferriter, Alice Goldfarb, and Olivia Dorsey for meeting with Lauren bi-weekly to share ideas, ask questions, and collaboratively learn. There is extensive hidden labor in keeping an initiative like this on track, and Jaime Mears kindly helped us navigate the process. Finally, thank you to our fellow researchers in-residence. Along with their exciting projects, Lincoln Mullen’s always astute assessments of the state of the field and Andromeda Yelton’s careful and critical use of machine learning helped us think bigger and broader about the impact of this initiative. Thank you to everyone involved for letting us be a part of a culture committed to collaboration, openness, experimentation, innovation, and most importantly, support and kindness.