Re: [libdv-dev] Possible bug encoding interlaced material?

[Jonathan W. <jw...@ph...>]

> > >   In general, I agree with the way you did your experiments but:
> > > 
> > >     1. You shouldn't be doing deinterlancing. There are very few
> > >        tools for that to handle PAL chroma issues correctly.
> > 
> > Um, the only deinterlacing I did was after the (interlaced) DV frame had
> > been extracted to a PNG file.  This (as mentioned on the web page) used
> > ffmeg's decoder via mplayer.  The resulting extracted PNG frame was still
> > interlaced, so any PAL chroma issues should have been taken care of by the
> > DV decoder used.  
> 
>   Nope. PNG uses RGB color space, while PAL DV operates in YUV420, thus
>   you need a conversion. And the conversion of PAL chroma is a picky
>   thing.

Ok, to eliminate any possibility (I think) that what we're seeing is related
to chroma conversion, I redid some tests over the weekend using YUV-planar
as an intermediate format and mjpegtools to handle the chroma conversions.
Thanks to Steven Schultz for his helpful tips.  Details of these revised
tests including command lines can be found towards the bottom of

  http://www.atrad.com.au/~jwoithe/mpeg/pantest/

Included in the description are links to the files created.  The practical
upshot is that the field shadows appear in the output from these tests in
exactly the same was as they did when mplayer was used to extract the
frames.  Unless there has been another error in the test method, this would
indicate that the shadowing effects are not related to chroma sampling
effects.  Note that I also repeated the test on libdv-encoded footage,
again with the same results as obtained previously.

>   ... the correct sequence of steps to get a truly telling result would
>   be:
>      
>      1. Extract to YUV4MPEG (make sure it's labeled as PAL).
>      2. Use the deinterlacer from the mjpeg tools.

This is what I have done (I think).

> > The other thing to note is that these artifacts are not present in the raw
> > footage.
> 
>   Interesting. Could you post a sample of that raw footage somewhere for
>   me to play with ?

http://www.atrad.com.au/~jwoithe/mpeg/pantest/straight_copy.mov

For completeness it's also linked as the source for straight_copy.m2v at the
top of

  http://www.atrad.com.au/~jwoithe/mpeg/pantest/

I didn't get a chance to use ffmpeg's DV decoder in a DV->MPEG2 over the
weekend; I'll try to get to that during the week.  Suffice to say that even
when my current libdv-based DV decoder is used as described in a previous
post, straight_copy.mov shows no artifacts on the DVD player (or in
extracted frames on the PC) whereas the same source encoded with libdv
before DV->MPEG2 conversion does.

Regards
  jonathan