ACTION is a set of analytical tools for cinematic information retrieval. It is the product of an NEH Digital Humanities Start-Up Grant awarded to Prof. Michael Casey and Prof. Mark Williams at Dartmouth, with the support of Dartmouth post-doc Tom Stoll.

From the ACTION website:

“Audio-visual Cinematic Toolkit for Interaction, Organization, and Navigation

ACTION provides a work bench to study such features in combination with machine learning methods to yield latent stylistic patterns distributed among films and directors. As such, ACTION is a platform for researching new methodologies in the study of film and media history.

The platform allows such features as access to and analysis of low-level frame-by-frame data, automated segmentation and clustering of this data, audio analysis for soundtracks, and other content analysis tools. ACTION provides Python tools to support research and development, and depends on the Bregman Python Toolkit. OpenCV 2.4 compiled with Python bindings is required for analysis.

Co-principle investigators: Michael Casey (Bregman) and Mark Williams (Film and Media Studies). Postdoctoral researcher: Tom Stoll (Bregman). Sponsored by the National Endowment for the Humanities (NEH), Office of Digital Humanities (ODH) Grant #HD-51394-11.”