[MPlayer-dev-eng] New inverse-telecine filter

[D Richard Felker III]

On Thu, Dec 04, 2003 at 08:22:35AM +0000, Adam Rice wrote:
> Zoltan Hidvegi wrote:
> >This is really bad, I do not know how can you get a good 24fps output
> >of this.  Variable rate seems to be more sutable for this, but my
> >filter still produces an acceptable result.
> 
> I think these tests are a little unfair, as the scenes are clearly 
> supposed to look grungy. Someone obviously went to a lot of trouble to 
> make them look like that, and if you somehow managed to clean it up 
> you'd be doing them a great disservice.

No one attempted to create combing. Combing doesn't exist on a TV
screen; it's an artefact of displaying pairs of fields which don't
belong together at the same moment in time, rather than during
alternate cycles of an interlaced refresh.

My goal, in doing inverse telecine, is twofold:

1) To recover all progressive sequences (24fps or 30fps) as actual
   progressive frames, so that they can be displayed correctly on a
   progressive-scan device.

2) To produce a progressive video output that looks as close as
   possible to what the original interlaced video would look like on
   an interlaced display, in the cases where perfect progressive
   frames cannot be recovered.

> I think in this circumstance all I'd expect from a de-telecine filter is 
> to produce output at the specified framerate that resembles the input, 
> and throw away as little information as possible.

Yeah, this is basically it.

Rich