Hello

y4mstabilizer works best with 444 progressive material.  This is fine
for making DVDs, since set-top players are supposed to interlace
material if required.  I have had good results with a pipeline like:

        ... | yuvdeinterlace | \
        y4mscaler -v 0 -O sar=src -O chromass=444 | \
        y4mstabilizer -a 0.8 | \
        y4mscaler -v 0 -O sar=src -O chromass=420_MPEG2 | \
        ... | mpeg2enc ...

(Yes, I know, it would be nice if it worked better with 411 or 420
interlaced material.  Life is short and I am very busy.  Sorry.)

Hmmm... You have given the jitter *amplitude* (12 pixels) but not the
jitter *frequency* (how often they occur) or duty cycle (how much of
the time is the image shifted).

How long do the jitters last?  Are they only a frame or two?  or do they
last several seconds?  Best results occur if the input shift is of short
duration, and quickly moves back to the original position.

The important parameter for you to vary would be the -a <alpha> value.
The higher the alpha value, the more "viscous" the output movement
becomes, resisting input movement.  Experiment to see what works best.

The lower the jitter frequency, and the higher the duty cycle, the
bigger alpha value will be required.  However, that also will resist the
intentional camera panning.

The default alpha value is 0.95, which is quite high.  I would start out
with an alpha of 0.8, and vary it from there.

I would not vary the -s <search radius> value unless the stabilizer has
difficulty tracking the motion.  Typically it is only useful to raise it
from the default 48.  The only possible reason to lower it would be to
speed it up, but y4mstabilizer is computationally insignificant next to
yuvdeinterlace.

Good luck.

Jim Macropol

On Tue, 2005-06-21 at 17:42 +1200, E.Chalaron wrote:
> Hi all
> 
> Question about y4mstabilizer.
> I have a reel here that is REALLY old.
> once reshot it does show some jittering (worned out brearing in the movie 
> camera itself ??).
> I have calculated about a dozen pixels for this up/down frequency.
> I want to compensate this but not the pannings that the cameraman did.
> Is there a way to do it properly ?
> My starting point would be to use -s 12, but any hints are welcome.
> Thanks
> E
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
> from IBM. Find simple to follow Roadmaps, straightforward articles,
> informative Webcasts and more! Get everything you need to get up to
> speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
> _______________________________________________
> Mjpeg-users mailing list
> Mjpeg-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mjpeg-users
-- 
J Macropol <[EMAIL PROTECTED]>



-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
Mjpeg-users mailing list
Mjpeg-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mjpeg-users

Reply via email to