Among a total of 32 new patent applications by Apple Inc. (NASDAQ:AAPL) published on Thursday, May 8th, 2014 by the United States Patent & Trademark Office (USPTO), one relates to a novel method of video editing using pixels in the actual video itself as the “handles” for accessing and editing the underlying assets making up that specific part of the video image. This invention would make video editing far faster, smoother, easier, and more intuitive than the current process of searching painstakingly through assets to try to determine which affect a specific part of the image.
Currently, in computer generated graphics, the pixels that appear in the film are caused by a web of underlying “illustration assets” and “objects.” These assets not only control the appearance of the image that the viewer sees, but also determine how it moves (following a programmed curve or line, for example), how it is positioned relative to other computer generated imagery, and so on.
According to the patent application, this makes editing specific images a time consuming and painstaking task. For example, moving or removing a single computer-generated tree involves finding the assets and objects used to generate it and “place” it within the final image. All subsidiary assets and objects must then also be found. All of these objects and assets are then individually edited until the desired positioning of the tree is achieved within the image. This “cutting room” work is lengthy, expensive, and prone to generating errors which must be fixed.
However, “Mapping Pixels to Underlying Assets in Computer Graphics,” the patent idea currently under consideration, changes the editing approach radically. In the approach described, the actual pixels of the image become linked to the assets and objects generating them, and can be used as a direct interface to gain quick, easy, intuitive editing control over those unseen layers of compiled data.
In effect, the process described, if successful, would be something like how a word processor currently works. Font, line spacing, justified margins, tabs, and so on are adjusted directly through the final, readable document – it is no longer necessary to go into HTML and rewrite the code to change, for example, font size and color. The program generates HTML code automatically based on choices at the document level.
In a similar way, the intricate complexity of video object, asset, and hierarchy data would be immensely streamlined if the pixels of the image themselves could be used to directly interface with it. The functionality would not be precisely the same as word processing, obviously, but the comparison is handy for illustrating how giving direct access to complex programming through the final product (typed page or pixelated video image) makes editing far faster and easier.
The patent application, #2014126888, was originally filed on November 2nd, 2012. Its publication now indicates that it is probably close to being granted. The assignee, naturally enough, is Apple Inc. of Cupertino, CA. Only a single inventor is listed for “Mapping Pixels to Underlying Assets in Computer Graphics,” Brian E. Goldberg of San Francisco, California.
