> Justin Randall wrote:
> > I guess that filmdeint could be used first in the filter chain, with
> > filmdint=dint_thres=256 to ensure that it doesn't try to deinterlace
> > any frame it can't match, then pass that output to mcdeint for getting
> > any remaining interlaced frames. I wouldn't recommend this though.
> > For one last try at the inverse telecline method you could try this:
> >
> > mencoder -vf pullup,softskip,pp=lb,hqdn3d=3:2:3:3 -ofps 24000/1001
> >
> > This should be good for animated sources, although pp=lb might cause
> > some ghosting. If you want to crop and/or scale, put those filters
> > after the denoiser. Example: mencoder -vf
> > pullup,softskip,pp=lb,hqdn3d=3:2:3:3,crop=704:480:0:0,scale=640:480
> > -sws 9 -ofps 24000/1001
>
> I find that adding 'harddup' as the last element in the filter
> chain prevents sync errors and never seems to hurt things.
Oops. Kind of wrote that without thinking too hard... It's correct,
except that the sync errors are rather occult. They would not appear
in the output container, but WILL appear if you then try to transcode
the output later on. Not sure what that's about, but it is solved
by adding harddup.
_______________________________________________
MEncoder-users mailing list
MEncoder-users@mplayerhq.hu
https://lists.mplayerhq.hu/mailman/listinfo/mencoder-users
No comments:
Post a Comment