[MPlayer-dev-eng] Re: More on inverse telecine!

[D Richard Felker III]

On Mon, Apr 14, 2003 at 09:13:20AM +0300, Andriy N. Gritsenko wrote:
>     Hi, D Richard Felker III!
> 
> Sometime (on Sunday, April 13 at 20:30) I've received something...
> [.......]
> >One dead giveaway that a frame is part of the telecine pattern is when
> >the max even-odd is massively smaller than max odd-even. The
> >difference is often as significant as 80 vs. 3000!
> 
>     When I wrote my dint filter then I've started from comparison with
> previous frame too. But then I've found it gives me not the best results.

Yes, but there are many ways of comparing frames, and some are not
useful at all for removing telecine.

> Anyway I don't know what your sources are but on two DVDs I've encoded
> were no standard telecine at all - somewhere they seemed as 30fps source,
> somewhere as 24fps with telecine, somewhere as source with about 60% of
> interlaced frames. The best results I've got with pp=lb,dint=0.2:0.002
> but anyway there are still interlaced frames and some good frames was
> skipped too. :(  I still wonder if there is a way to really deinterlace
> such kind of sources...

Yep, there definitely is. I've largely had success with SE:Lain, using
vf_detc, and Lain has some really weird patterns!! Perhaps this is the
DVD you're referring to? :)

BTW, how do you use vf_dint?? I've never been able to get it to do
anything...

Rich