Techniques for Managing Variable Frame Rate Images

Abstract

Much has been presented regarding the history of the selection of frame rates for film and television, and the technical necessity for the television rate and the artistic merits of a slower rate for film have also been widely discussed. Given that film formed the traditional source for much of the high quality content for television, conversion via the classic 2:3 pull-down technique is well understood. Recently, however, cinematographers have increasingly looked to video acquisition as a cost effective option for the making of films as well as television programming, and an increasing number of films being shown at the major film festivals were acquired on video. While some of these were shot at traditional 60 field interlace (60i), many cinematographers resorted to the use of 25 frame 625 line (nominally PAL) equipment to attempt to answer their requirements. Neither 60i nor 25P convert seamlessly to 24P. The 60i material suffers from several short-comings (e.g. interlace artifacts, to name just one issue), and the 25/24 (a not insignificant 4% difference) between the speeds brings no end of complications (e.g. audio pitch and timing difficulties, to name just one issue). To satisfy these users video equipment operating at a native 24 frame progressive rate has been developed in both high definition and standard definition formats, giving the filmmaker accurate and immediate feedback about what a scene will look and sound like when finished. — This has allowed a significant portion of episodic television program production to adopt a 24 frame progressively scanned video format wherein the production process virtually replicates film. Two factors have evolved from this: a desire on the part of the production community for a variable frame rate camcorder to more fully mimic the flexibility of film cameras, and a desire on the part of a larger market for access to 24 frame standard definition video acquisition. Numerous presentations have been given on a variable frame rate production camcorder that has been well received and widely adopted, and a very compact and cost effective 60i/30p/24p frame DV camcorder has been brought to market that uniquely carries 24 frame progressive information in a 60i format. Now, as more main-stream DVCPRO50 professional equipment enters the marketplace and spans the gamut of equipment levels needed by the industry, it is increasingly important for users to understand how these devices actually process the video frames, store them in standard television frame rate video formats, and enable post production recovery and manipulation in the “native” frame rate. This paper describes and compares “standard” 2:3 pull-down, a new advanced 2:3:3:2 pull down that allows full recovery of the original progressive material when stored in an easily viewable interlaced format, as well as the techniques within the variable frame rate process that isolate the recording mechanism and its fixed frame rate from the image capture section and its user selectable frame rate with a kind of n:60 pull down. Many users have had difficulty understanding how complete the analogy is between an image on film and an image recorded using these techniques, and this is explained in some detail. In addition, one can extrapolate that future products can use techniques like increased memory to further isolate the respective frame rates.