How Will We Preserve the Movies of Tomorrow? Technical and Ethical Considerations

In this new age of computational imaging, volumetric video, performance capture, visual effects, and a myriad of other technologies we are creating images that were never captured in the conventional sense. These new technologies offer an almost infinite range of creative options once a production is over: camera angles can be modified; virtual lenses can deepen or tighten the depth of field, expand or compress perspective, and even mimic the distortion of vintage lenses; actors can be replaced, or even brought back to life decades after they passed. On top of this imaging revolution, the rapidly evolving field of image display means that a final version for today may not be optimal or even adequate for the screens of tomorrow, whether due to increased color gamut, dynamic range, spatial resolution, size, aspect ratio, or frame rate. These issues bring up technical, financial, and ethical considerations about what to conserve, in which form, and for how long, with potentially conflicting priorities between the preservation of the original artistic intent and a content owner's desire to maximize the revenue and/or lifespan of a particular asset. For example, if a character is a computer-generated avatar of a performance given through motion capture, should we archive, on top of the on-screen performance, all the motion capture files, from the initial video capture to the computerized animation to the character models to the final render before it was composited into the shot? Could we, and should we, leave that door open for future generations to change characters in the finished product, eventually without the original creator's consent? On the other hand, if we do not conserve the original animation, do we hamper a creator's freedom to upgrade and refresh an asset when new technologies arise in the coming years? At what cost do all these decisions come? In this article, the author proposes a new definition for 'image sequence-based works of art,' to replace the traditional definition of 'film,' and details workflow options that strike the right balance between the protection of creative intent and future proofing of assets, combining technical, ethical, and philosophical considerations. With all the advances we have witnessed in the past 10 years, and what is to come in the next decade, the author believes there is an urgency to answering these questions and standardizing workflows before what we do today has an unalterable impact for future generations. © 2002 Society of Motion Picture and Television Engineers, Inc.