Browse Prior Art Database

Video Frame-Rate Reduction Technique

IP.com Disclosure Number: IPCOM000112942D
Original Publication Date: 1994-Jun-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 8 page(s) / 255K

Publishing Venue

IBM

Related People

Moller, CHL: AUTHOR

Abstract

Given an ordered set of digitised video frames representing an animated image, the described technique will create a second ordered set of such frames that, if displayed sequentially, will either correct for a sampling rate slower than the display rate, or create the appearance of slowed motion.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 22% of the total text.

Video Frame-Rate Reduction Technique

      Given an ordered set of digitised video frames representing an
animated image, the described technique will create a second ordered
set of such frames that, if displayed sequentially, will either
correct for a sampling rate slower than the display rate, or create
the appearance of slowed motion.

      The appearance of motion on a video display is done by showing
sequences of static images in rapid succession.  Typically, the
sequence of static images is obtained either by sampling and
digitising a scene using a video camera of some sort, or by rendering
a succession of synthetic images using appropriate computer-graphic
techniques.  In the former case, the sequence has been captured at a
specific sampling rate; in the latter case, a specific real-time
interval has been assumed between the computation of successive
frames.

      The sequence, having been sampled or created at a specific
rate, can only be replayed accurately if the successive images are
displayed at that same rate.  It is possible, however, for the sample
rate to be different than the display rate.  For example, television
images are typically sampled at 30 frames per second (fps), while
computer monitors usually refresh at a different rate such as 44 fps.
Under these circumstances, one second of a source sequence,
comprising 30 frames, would be displayed in 30/44 seconds, resulting
in a apparent acceleration of the motion represented.

      Under other circumstances, it may be desirable to play an
animation at a speed slower than that which it was captured.  Given
equivalent capture and display rates, this is essentially the same
problem as above:  increasing the number of frames to be displayed
while retaining the content of the original animation.

      The problem is thus to "stretch" the number of source frames to
a larger number of resultant frames.  From the previous example, the
original 30 source frames must somehow be used to create 44 resultant
frames, each sequence of frames representing one second of real time
in its own domain.  Alternatively, given an animation sampled at a
given rate and a desire to display that animation slowed by some
arbitrary factor on equipment using the same refresh rate, it is
necessary to increase, by that same factor, the number of frames
available.

      A simple solution to the rate-adjustment problem is through the
use of a frame buffer, allowing the sequence source to update pixels
in the frame buffer at one rate, while allowing the display system to
sample the same frame buffer at another rate.  This mechanism,
however, results displayed frames that consist of parts of two
sequential source frames.  This in turn results in various kinds of
image distortion such as elongation or foreshortening of objects as
they move.  Under some circumstances, moving single-pixel-thick
horizontal lines can flicker in and out of visibility or vanish
entirely.

   ...