Media Digitization Project

Ben Olander with Jessica Feldman (Communication, Media, and Culture), Media Digitization Project 

The Media Digitization Project harnessed new digital tools and technologies in an effort to digitize historical and deteriorating 8mm and 16mm film at an unprecedented quality, in a scaleable manner and at an affordable cost. Using new larger format camera sensors, low cost dedicated Machine Learning (ML)/Artificial Intelligence (AI) processing hardware and custom built software for quality assurance, automation and restoration, this film digitization project can be a powerful tool for preserving priceless visual testimony of 20th century life. The existence of an ad hoc global collection of amateur, and independent film (shot in 8mm and 16mm) has been a critical component of the development of civic society in the 20th century, and it is currently under threat of vanishing due to the decomposition of celluloid materials used in film-making, or the extremely low quality VHS transfers done in the past. While there are currently options for digitizing many of these works, they are often low-quality, simply swap one degrading media format for another, or are excessively expensive. There is a critical need for low cost, high quality digitization of these irreplaceable historical and cultural documents before the underlying celluloid based media dissolves permanently. This project protoyped new low cost hardware options, and paired them with new ML/AI Machine Vision software solutions to radically lower the barrier of entry, and dramatically increase the quality of preserved film media.