As the Raspberry Pi Foundation (RPF) has worked to make computing more accessible, it has helped pioneer new ways of using technology. We've seen the versatile, board-based Raspberry Pi enabling everything from robotic bartenders to doggie treat dispensers. The latest project featuring the Pi comes from Matthew Epler, whose Pi-powered Kinogarph digitizes old film stock at a fraction of the cost of conventional off-the-shelf systems.
Epler's thesis project for NYU's Interactive Telecommunications Program, the Kinograph uses software to stabilize a series of captured images and extract optical sound. In addition to a Raspberry Pi, the system also requires a consumer-grade digital camera, a few 3D-printed parts, such as rollers that can take different sized film, and a few components that can be cheaply sourced over the internet.
This means that a decent DSLR will be the most expensive part of the setup, with the other components to add up to around $1,200, although Epler is hoping to get this down to under $1,000. As a point of comparison, the cost to digitize 50 films on reels at a film lab would cost around $480,000, while buying an off-the-shelf commercial equivalent would cost over $150,000. As Epler explains, "the cost of digitization is prohibitive for anybody with more than a few films." This is as true for individuals with private collections as it is for libraries, museums and other public institutions.
Capable of digitizing 35, 16, and 8 mm films at high-definition resolutions with sound, the Kinograph makes preserving our cultural heritage much more affordable. The image capture system uses openFrameworks, an Arduino, and embedded Linux on a Raspberry Pi, while OpenCV, Processing, and AEO Sound applications are used for image and audio processing, with batch processing of crop, rotation and color correction all possible. Everything utilized in the Kinograph is open-source and available online.
Epler's philosophy behind the project speaks to a broader perspective and says, "Cultural memory should not be dependent on money or technology." It's only fitting that he chose to use a Raspberry Pi for the project, as their perspective on learning and computing is similar.
The RPF estimates that more than 90 percent of all silent movies, and around 50 percent of the films with audio made before 1950 have been lost to the ages. Epler is striving to recapture our past with the powerful, yet cheap and accessible, technology of our present.
The following video gives a brief overview of the Kinograph.
Sources: Matthew Epler, Raspberry Pi
