[Mp4-tech] Deriving the Chroma motion vectors from Luma motion vectors.

[Satendra]

Hi,
My same old querry as answered .
--------------------------
Satendra
From: *Soumen Kumar Dey *
*As you know Chroma Motion vectors are half of Luma*
*motion vectors. *
*mvc = mvluma/2;*
*But chroma motion vectors are calculated for*
*1/8 pixel accuracy in respect of luma's 1/4 pixel accuracy.*
*OriMVc = mvc*2 = mvluma;*
**
*[2 = (1/4)/(1/8)]*
*While deriving Integer position of chroma*
*MVCFrac = (OriMVc&7);*
*MVCInt = (OriMVc>>3);*
**
*I think it is sufficient for ur question.
*
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
This section is being clarified in the new corrigendum.
In this case, it is best to pay more attention to the equations than to
the words.
Note that, for 4:2:0 video, the units of the chroma motion vectors are
1/8 (chroma) sample units and the units of the luma motion vectors are
1/4 (luma) sample units.
Therefore if you set the value of a chroma motion vector equal to the
value of a luma motion vector, a "virtual" division by 2 is occurring.
E.g., a value of 3 for a luma motion vector indicates 3/4 luma
full-sample units, and a value of 3 for a chroma motion vector indicates
3/8 chroma full-sample units; so although the values of these two
vectors are equal to each other, the chroma motion vector has half the
interpreted amplitude as that of the luma motion vector (in the domain
of its use).
Best Regards,
Gary Sullivan
+> -----Original Message-----
+> From: Satendra [mailto:spsatendra gmail.com]
+> To: jvt-experts mail.imtc.org
+> Subject: [jvt-experts] chroma MV prediction
+>
+> Hi all,
+>
+> Its a very basic question, still please tell me.
+>
+> can anyone explain me how chroma MVs are calculated in decoder side.
+> IN section 8.4.1.4 Derivation for chroma MVs it is written,
+>
+> " If the current macroblock is a frame macroblock, the horizontal and
+> vertical components of the chroma motion vector mvCLX are derived by
+> multiplying the corresponding components of luma motion
+> vector mvLX by
+> 2, through mapping one-quarter sample mvLX units to one-eighth sample
+> mvCLX units."
+>
+> and eqn 8-174/5 states
+>
+> mvCLX[0/1] = mvLX[0/1]
+>
+> I am not able to understand the words in the above para. where are we
+> multiplying the luma MVs?
+>
+> Also in the field coding mode, what the parity conditions signify.
+>
+>
+> Thanks
+>
+> Satendra
+
On 2/23/06, Prasanna Sethuraman <prasanna.sethuraman patni.com> wrote:
>
>
>
> Dear all,
>
>
>
> The JVT draft for H.264 specifies that the chroma motion vectors should be
> derived from the estimated Luma motion vectors by multiplying with 2:
>
> "If the current macroblock is a frame macroblock, the horizontal and
> vertical components of the chroma motion
>
> vector mvCLX are derived by multiplying the corresponding components of
> luma motion vector mvLX by 2,
>
> through mapping one-quarter sample mvLX units to one-eighth sample mvCLX
> units
>
> mvCLX[ 0 ] = mvLX[ 0 ]      (8-174)
>
> mvCLX[ 1 ] = mvLX[ 1 ]      (8-175)"
>
>
>
> But logically, since the chroma components have only half the number of
> samples compared to luma, should not we be diving the luma motion vectors by
> 2? Or am I missing something?
>
>
>
> --
>
> Prasanna.
>
>
>
> http://www.patni.com
> World-Wide Partnerships. World-Class Solutions.
> _____________________________________________________________________
>
> This e-mail message may contain proprietary, confidential or legally
> privileged information for the sole use of the person or entity to whom this
> message was originally addressed. Any review, e-transmission dissemination
> or other use of or taking of any action in reliance upon this information by
> persons or entities other than the intended recipient is prohibited. If you
> have received this e-mail in error kindly delete this e-mail from your
> records. If it appears that this mail has been forwarded to you without
> proper authority, please notify us immediately at netadmin patni.com and
> delete this mail.
> _____________________________________________________________________
> _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include [audio,
> [video], [systems], [general] or another apppropriate identifier to indicate
> the type of question you have.
>
> Note: Conduct on the mailing list is subject to the Antitrust guidelines
> found at
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>
>

--
-----------------------------------------------------------------------------------------------------------------------------------
"We all agree on the necessity of compromise. We just can't agree on when
it's necessary to compromise."  ------Larry Wall
-----------------------------------------------------------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20060224/667be73e/attachment-0001.html