Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
SYSTEM AND METHOD FOR AUTOMATIC ZOOM
SPECIFICATION
FIELD OF THE INVENTION
[0001] The present invention relates to digital video signals. More
specifically, the present invention relates to digital video content ratios in
a digital
video streain.
BACKGROUND OF THE INVENTION
[0002] Today broadcast video conteilt is available in both a 4:3 ratio and
16:9
ratio formats. Typically, standard definition ("SD") broadcast video is sent
in 4:3 and
high definition ("HD") broadcast video is sent in 16:9. However, sometimes
television
broadcasters send HD video signals containing programs that were originally
filmed
in the 4:3 format ("Ratio Mismatch"). In this instance, the HD broadcaster
must add
vertical bars ("Pillar Bars") to the left and right sides of the picture to
fill the entire
16:9 HD display area. These Pillar Bars are typically black in color, but also
may be
gray, or some other color or graphic.
[0003] In the case of a Ratio Mismatch where 16:9 content is sent in a 4:3
video stream, horizontal Pilar Bars are added to the top and bottom of the
picture to
fill the entire 4:3 display area. Horizontal Pillar Bars are also conunonly
referred to as
"Letterbox Bars". As used herein the term "Pillar Bars" is used with reference
to both
vertical (left and right) bars and horizontal (top and bottom) bars.
[0004] To compensate for the above-described Ratio Mismatch, 16:9 HD
televisions/displays and set top boxes typically have a manual "Zoom"
function. The
Zoom function is usually implemented via a button on an associated remote
control to
(i) stretch the image horizontally to remove the Pillar Bars thereby
compromising the
aspect ratio, or (ii) stretch the image both horizontally and vertically
thereby filling
the display and preserving aspect ratio, while sacrificing a portion of the
top and
1
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
bottom of the image. (The Zoom feature can also prevent uneven display
degradation
on plasma monitors.) This manual Zoom process is totally controlled by the end
user
and can be imprecise, inconsistent, and time consuming because the end user is
required to press a button(s) on the remote control and step/cycle through
Zoom
states. Further, when the video content changes ratio (the Pillar Bars
disappear or
change size, for example), the end user will again have to re-adjust the Zoom,
either
"in" or "out" manually. In addition, if the video content changes ratio when
the end
user is performing their manual Zoom, the end user will become highly
frustrated, as
the end user attempts to Zoom to compensate for a ratio which has since
passed. An
end user's only other option is to select a static zoom setting, which may
under or over
Zoom depending on the Ratio Mismatched video content size.
[0005] Thus, what is needed is a system and method which provides a more
consistent and efficient process for zooming "in" and "out" a video content
stream to
adjust for Ratio Mismatches.
2
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
SUMMARY OF rNVENTION
[ 0006] An object of the present invention is to provide for automatic zooming
"in" and "out" of a video content stream to adjust for Ratio Mismatches.
[0007] In order to achieve this objective, as well as others which will become
apparent in the disclosure below, the present invention provides for an
automatic
zoom system and method for automatically zooming "in" and "out" a video
content
streain to adjust for Ratio Mismatches. The present invention is particularly
applicable
to 4:3 video content delivered in a 16:9 video content stream, which is often
found in
HD video content streams.
[0008] In an exemplary embodiment of the present invention, the end user
selects the desired Auto-Zoom setting on their television, video display, or
video set
top box at the time of initial setup. Such setting would preferably be a one
dimensional, horizontal or vertical zoom only (for vertical and horizontal
Pillar Bars,
respectively); or a two dimensional zoom, intended to preserve aspect ratio.
The
present invention then scans for the presence of Pillar Bars. When pillar bars
are
detected by the present invention, after waiting a pre-determined period to
confirm the
presence and size of the Pillar Bars, the present invention slowly initiates a
stretching
of the video content stream to eliminate the Pillar Bars in accordance with
the Auto-
Zoom setting ( above).
[0009] When the video content stream changes back to without Pillar Bars, the
present invention initiates a slow zooming out to a full pixel representation
of the
video content stream. Further, the present invention constantly determines the
presence of Pillar Bars, so that as the video content streain alternates
between a full
16:9 presentation and a Ratio Mismatch, the present invention automatically
zooms
the video content stream "in" and "out" as needed.
[0010] Thus, the present system and method provides for automatically
zooming "in" and "out" a video content stream to adjust for Ratio Mismatches.
3
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
BRIEF DESCRIPTION OF THE DRAWINGS
[0011] For a complete understanding of the present invention and the
advantages thereof, reference is now made to the following description taken
in
conjunction with the accompanying drawings in which like reference numbers
indicate like features, components and method steps, and wherein:
[0012] FIG. l(a) is an illustration of 4:3 content transmitted in a 16:9 video
bit
stream;
[0013] FIG. 1(b) is an illustration of 4:3 content transmitted in a 16:9 video
bit
stream with a matrix overlay in accordance with an exemplary embodiment of the
present invention;
[0014] FIG. 2(a) is an illustration of 4:3 content transmitted in a 16:9 video
bit
stream after automatic zoom in accordance with an exemplary embodiment of the
present invention;
[0015] FIG. 2(b) is an illustration of the matrix overlay over 4:3 content
transmitted in a 16:9 video bit stream after automatic zoom in accordance with
an
exemplary embodiment of the present invention;
[0016] FIG. 3 is an illustration of the basic process flow of the method for
automatically zooming "in" and "out" a video content stream to adjust for
Ratio
Mismatches in accordance with an exemplary embodiment of the present
invention;
[0017] FIG. 4(a) is an illustration of the viewable area of 4:3 content
transmitted in a 16:9 video bit stream in a display after automatic zoom in
accordance
with an exemplary embodiment of the present invention for use with FIG. 4(b);
and
[0018] FIG. 4(b) is an illustration of the over-scan regions of 4:3 content
4
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
transmitted in a 16:9 video bit stream after automatic zoom in accordance with
another exemplary embodiment of the present invention.
DESCRIPTION OF A PRESENTLY PREFERRED EMBODIMENT
[0019] FIG 1.( a) is an illustration of an instance of a 16:9 video stream
where
the broadcaster has sent 4:3 video content and has used left and right Pillar
Bars to fill
the remainder of the pixel space ("Ratio Mismatch"). This type of transmission
is
most typically used when a HD broadcaster is transmitting 4:3 video content in
their
16:9 video content stream.
[0020] The present invention provides for a system and method for
automatically zooming "in" and "out" a video content stream to adjust for
Ratio
Mismatches.
[0021] Referring to FIG. L(b) and FIG. 3, the system and method ofthe
present invention begin by (a) defining an instance of a video content stream
into a
matrix of quadrants, in a matrix component, in step 302, as can be seen in the
exemplary instance in of FIG. 1.(b). A 10x20 matrix is used in FIG. 1(b)
simply as an
illustrative example. Preferably a 72x48 matrix is used. However, any matrix
size, up
to and including the full resolution of the video source, can be used.
[0022] Next, the present invention (b) measures a characteristic in each
quadrant of the matrix of that instance, in a measuring component, in step
304. Such
characteristic may include luminance and/or chrominance, for example.
Preferably
luminance of the pixels in each quadrant is the characteristic measured in
step (b) 304
because by using luminance, the present invention can analyze Pillar Bars of
any
color or pattern, e.g., black, gray, graphics, etc.
[0023] The present invention then repeats steps (a) 302 and (b) 304 for a
5
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
number of iterations, to (c) determine which quadrants have the characteristic
in a
steady state condition, in a determining component, in step 306. In this
exemplary
embodiment, steps (a) 302 and (b) 304 are repeated for 30 cycles with a one
(1)
second interval between each cycle. However, there are numerous variants for
the
number of cycles and interval therebetween, which can be determined by one of
ordinary skill in the art.
[0024] After quadrants which are in a steady state are determined, in step (d)
306, the present invention uses that steady state quadrant determination to
(d)
determine the size of the Pillar Bars in the video content stream, in a size
component,
in step 308. In an exemplary embodiment, in step (d) 308, the present
invention
evaluates each row in the matrix for quadrants determined in step (c) 304 to
be in a
steady state condition. After analyzing each row, the present invention
derives the
lowest number of quadrants in a steady state condition across all rows of the
matrix
(also can be described as "number of columns in the matrix with all quadrants
in a
steady state condition"). The resultant number equals the width of the Pillar
Bars (left
and right). Referring to FIG. 1(b), the lowest number of quadrants in a steady
state
condition across all rows of the matrix would be 4.
[0025] For efficiency purposes, the present invention may simply analysis half
of each row to determine the width of one Pillar Bar. For example, if only the
first
half of each row is analyzed for steady state quadrants, the width of the left
Pillar Bar
can be determined. In this case, the present invention assumes the right
Pillar Bar is of
comparable width and, thus, will still arrive at the width of the Pillar Bars.
[0026] After determining the resultant lowest number of quadrants in a steady
state condition across all rows of the matrix, the present invention
horizontally zooms
in the video content stream to an amount equal to the size of the Pillar Bars,
in an
automatic zoom component, in step 310. The resultant "zooming in" can be seen
in
FIG. 2(a). The type of Auto-Zoom (one-dimensional or two dimensional) used can
be
end user defined as described above.
6
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
[0027] After completing step (d) 308, the present invention return to step (a)
302 and begins the process again. Preferably, the time interval between the
completion of step (d) 306 and returning to step (a) 302 to begin the
inventive method
again is one (1) second. In this way, the present invention repeatedly
monitors the
size/presence of Pillar Bars. This allows the present invention to detect the
elimination of previously determined Pillar Bar (as described below), as well
detection of new wider Pillar Bars (where the lowest number of quadrants in a
steady
state condition across all rows of the matrix is higher than the previous
state) so the
present invention can automatically zoom "in" more. FIG. 2(b) illustrates the
recursive process described above as a matrix is still overlayed over an
instance of the
video content stream for repeated quadrant steady state condition analysis.
[0028] To account for a change in the video content back to a full 16:9 video
stream, the present invention provides monitoring to determine quadrants in
the above
determined Pillar Bars are no longer in a steady state condition. Preferably,
if at least
one quadrant in the above determined Pillar Bars are no longer in a steady
state
condition, the present invention determines that the Pillar Bars (or at least
the same
Pillar Bar(s)) do not exist. In that case, the present invention zooms the
video content
stream out to a full pixel representation.
[0029] In accordance with another exemplary embodiment for zooming out, in the
case that a television, set-top box, or other display device will not allow
for quadrant
steady state condition analysis of quadrants out of its display area due to
the inventive
zoom "in" (described above, where the Pillar Bars are completely eliminated)
("Truncated Display"), the present invention provides for an alternative
automatic
zooining "in" where the video content stream is zoomed "in" to an amount equal
to
the size of the Pillar Bars minus a small portion of the Pillar Bars. The
small portion
of the Pillar Bars are preferably approximately equal to the over-scan region
in most
display devices, as to keep this small portion of the Pillar Bars outside of
the view of
the end user but still in the display. FIG. 4(a) and 4(b) illustrate this
inventive
7
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
technique and the sinall portion of Pillar Bars in the over-scan region.
[0030] Using this small area of the Pillar Bars in the over-scan region, a
display device which cannot analyze quadrants which have been cropped due to
the
zoom, can use the small Pillar Bars portion to determine if at least one
steady state
quadrant in this small region is no longer in a steady state condition. In
that case, the
present invention assumes that the full 16:9 video content has returned to the
video
content stream and zooms the video content stream "out" to a full pixel
representation.
Thereafter, analysis beginning with step (a) 302 begins again.
[0031] Using the same inventive system and method described above, top and
bottom Pillar Bars can be determined and eliminated by zooming "in". In this
instance, step (d) would comprise determining the lowest number of quadrants
in a
steady state condition across all columns of the matrix (also can be described
as
"number of rows in the matrix with all quadrants in a steady state
condition"). This is
useful when presenting 16:9 content in a 4:3 video streain, such as playing a
16:9
DVD movie on your DVD player which is outputting a 4:3 video content stream.
Further, the inventive technique for Truncated Displays may also be employed
when
the present invention performs an automatic zoom "in" to remove a top and
bottom
Pillar Bars.
[0032] The present invention may also comprise an initial step, before step
(a)
302 of determining the current zoom state ("zoomed out" to full pixel
representation,
"zoomed in" at particular number of quadrants, etc.) to form a base zoom state
for the
initial zooming and subsequent zooming "in" and "out" by a zooming state
component.
[0033] Thus, the present system and method provides for automatically
zooming "in" and "out" a video content stream to adjust for Ratio Mismatches.
[0034] Although the invention has been described herein by reference to an
8
CA 02594304 2007-06-21
WO 2006/071956 PCT/US2005/047316
exemplary embodiment thereof, it will be understood that such embodiment is
susceptible of modification and variation without departing from the inventive
concepts disclosed. All such modifications and variations, therefore, are
intended to
be encompassed within the spirit and scope of the appended claims.
9