[opendtv] Re: which wide gamut (was: 1080P Question)
- From: Kilroy Hughes <Kilroy.Hughes@xxxxxxxxxxxxx>
- To: "opendtv@xxxxxxxxxxxxx" <opendtv@xxxxxxxxxxxxx>
- Date: Thu, 17 Sep 2009 15:54:28 +0000
The problem is that with negative R'G'B' coefficients, the encoded colors are
no longer limited to the area of the Rec 709 color primaries. You have the
ability to capture and encode some other color gamut using those arbitrary
vectors to map that 3 space. If you limit to the +, +, + quadrant and maximum
RGB lengths, your vectors define the space of your encoded colors. If you
allow vectors that are >1 or negative, then you can map any color space you
want regardless of the primaries.
So if you want to provide the display process that pallet maps the Y'CbCr
values with useful information it can use to scale the incoming gamut to a
"legal" 709 gamut or some other extended gamut, then the encoded content should
really provide some hints as to the color range that has actually been encoded;
not just the primaries that were abused in the process.
You could use the 709 primaries and encode the DLP/D-Cinema gamut, for
instance. The display process might eventually be able to guess what colors
were actually encoded after it sees enough "DLP" colors pop up, but it would be
more deterministic to tell it up front to it could do a stable gamut remap.
-Kilroy
[JS]" I am not sure that I understand you. Traditionally, wide color gamut
data would be related to
some non-standard set of color primaries. There are countless examples of such
ad-hoc and
real standards, the best known ones are probably AdobeRGB and XYZ. Obviously
one must
specify the co-ordinates of the 3 primary colors, or else the R',G',B' image
data is meaningless.
xvYCC image data (or any Y'CbCr data for that matter !) is always referred to
the Rec.709
(sRGB) color primaries, therefore for any Y'CbCr vector one can unambiguously
calculate
the XYZ or (x,y) or (u',v') co-ordinates. Thus the image itself contains
everything you need
to know about its color gamut. Just find the most extreme colors in the data.
Gamut boundary
data can only give you some more information about the potential gamut of
images, but you
can not know if every individual image will ever reach that boundary. I am not
sure how that
will help you to define the optimum rendering intent for any given (individual)
image ? "
From: opendtv-bounce@xxxxxxxxxxxxx [mailto:opendtv-bounce@xxxxxxxxxxxxx] On
Behalf Of Stessen, Jeroen
Sent: Thursday, September 17, 2009 6:31 AM
To: opendtv@xxxxxxxxxxxxx
Subject: [opendtv] Re: which wide gamut (was: 1080P Question)
Hello,
Kilroy Hughes:
Ø I am just focused on getting wide gamut encoded "right" at this point, but
the method HDMI uses to signal color boundaries might be a useful addition in
an SEI message (an optional/informative element in the H.264 bitstream).
I believe that the design of this metadata originates from Jurgen Stauder at
Thomson (dot net).
If you've got several weeks of spare time on your hands you may want to study
his papers.
Ø H.264 has specific VUI (Video Useability Information) parameters that are
pitched to the Display Process so it can implement the correct "render intent"
e.g. wide gamut. Only thing lacking is a method to signal "which wide?".
I am not sure that I understand you. Traditionally, wide color gamut data
would be related to
some non-standard set of color primaries. There are countless examples of such
ad-hoc and
real standards, the best known ones are probably AdobeRGB and XYZ. Obviously
one must
specify the co-ordinates of the 3 primary colors, or else the R',G',B' image
data is meaningless.
xvYCC image data (or any Y'CbCr data for that matter !) is always referred to
the Rec.709
(sRGB) color primaries, therefore for any Y'CbCr vector one can unambiguously
calculate
the XYZ or (x,y) or (u',v') co-ordinates. Thus the image itself contains
everything you need
to know about its color gamut. Just find the most extreme colors in the data.
Gamut boundary
data can only give you some more information about the potential gamut of
images, but you
can not know if every individual image will ever reach that boundary. I am not
sure how that
will help you to define the optimum rendering intent for any given (individual)
image ?
Ø That is a major improvement over MPEG-2, which was designed when the world
had one choice (Rec 601), so the bitstream didn't have to deal with these
options.
I am not sure that Rec.601 specifies any colorimetry, I thought that it only
specified the matrix
for R'G'B' to Y'CbCr and back. The actual meaning of R'G'B' co-ordinates was
standardized
in several other places, and that should not include NTSC 1953 anymore. Charles
Poynton says
that for all practical purposes the color gamut that everybody uses is
standardized in Rec.709.
xvYCC only extends on that, but in such a way that for the colors inside the
original gamut
nothing changes. From the point of view of a TV designer, dealing with
out-of-gamut colors is
not much different from dealing with abnormal settings of the user controls:
contrast,
brightness and saturation. You have to solve that without the benefit of
metadata too.
Groeten,
-- Jeroen
Jeroen H. Stessen
Specialist Picture Quality
Philips Consumer Lifestyle
Advanced Technology (Eindhoven)
High Tech Campus 37 - room 8.042
5656 AE Eindhoven - Nederland
Office mobex: (27) 99650
VoipBuster: +31.85.785.3611 (VoIP-in to PC)
Mobile 1: +31.6.1389.2492 (office gsm)
Mobile 2: +31.6.4468.0021 (personal gsm)
E-mail 1:
Jeroen.Stessen@xxxxxxxxxxx<mailto:Jeroen.Stessen@xxxxxxxxxxx>
E-mail 2:
Jeroen.Stessen@xxxxxxxxxx<mailto:Jeroen.Stessen@xxxxxxxxxx>
Website: www.consumer.philips.com<http://www.consumer.philips.com>
From: opendtv-bounce@xxxxxxxxxxxxx [mailto:opendtv-bounce@xxxxxxxxxxxxx] On
Behalf Of Kilroy Hughes
Sent: Wednesday, September 16, 2009 22:04
To: opendtv@xxxxxxxxxxxxx
Subject: [opendtv] Re: 1080P Question
I am just focused on getting wide gamut encoded "right" at this point, but the
method HDMI uses to signal color boundaries might be a useful addition in an
SEI message (an optional/informative element in the H.264 bitstream). H.264
has specific VUI (Video Useability Information) parameters that are pitched to
the Display Process so it can implement the correct "render intent" e.g. wide
gamut. Only thing lacking is a method to signal "which wide?".
That is a major improvement over MPEG-2, which was designed when the world had
one choice (Rec 601), so the bitstream didn't have to deal with these options.
________________________________
The information contained in this message may be confidential and legally
protected under applicable law. The message is intended solely for the
addressee(s). If you are not the intended recipient, you are hereby notified
that any use, forwarding, dissemination, or reproduction of this message is
strictly prohibited and may be unlawful. If you are not the intended recipient,
please contact the sender by return e-mail and destroy all copies of the
original message.
Other related posts:
