On 6 Nov 2004, scott wrote:
> That's interesting, I'm going to put some of my code in there instead of
> y4mshift. I want a specific vertical shift (im my case 0 or -1 or -2
Ah, sort of a "blend of features" from several programs ;)
> > y4mscaler can do that for you with the option "-O chromass=422" or
> > "-O chromass=444". Then to convert back to 420 before going into
> > the encoder you'd use "-O chromass=420_mpeg2"
> >
> That sounds easy :-)
It is easy :) Just add a "y4mscaler -O " before and after the
processing you want to do.
Adds a bit of overhead to the process (pipes are cheap but they're
not "free") but that can't be avoided unless one writes a monolithic
program.
> > The other thing which may make life simpler is to deinterlace the
> > material with 'yuvdeinterlace'. Deinterlaced supersampled (to 4:4:4)
> >
> Yes, it would be easier except that every 5 frames or so the field order
> is incorrect and needs to be swapped. I'll have a look at y4mscaler -S
> mode=LINESWITCH so see if I can make it only line switch on a function I have.
Hmm, but if you deinterlace does it matter (much) if the fields are
reversed? You'll still get the single blended progressive frame out.
What kind of capture process produces such a weird "jitter" or
field reversal? That sounds very strange.
Cheers,
Steven Schultz
-------------------------------------------------------
This SF.Net email is sponsored by:
Sybase ASE Linux Express Edition - download now for FREE
LinuxWorld Reader's Choice Award Winner for best database on Linux.
http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
_______________________________________________
Mjpeg-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/mjpeg-users
