[mythtv-users] Daddy, what's a commercial

Daniel Walton dwalton at cisco.com
Fri Jan 12 20:22:14 UTC 2007


On Fri, 12 Jan 2007, Michael Lynch wrote:

> > > If you are running svn you may also want to give the Experimental
> > > detection
> > > method a try.  It currently uses logo detection and blank frame
> > > detection but
> > > there are additional methods in the works.
> >
> >
> > The present system is certainly good enough for my purposes. I don't
> > think it will ever be 100% and we are probably reaching the point of
> > "diminishing returns" in that 100% more effort will result in at best
> > a percent or two of improvement in performance.

Maybe someday but I don't think we're there yet.  Adding audio analysis would be
a big plus as would some true pattern recognition for determining if a frame
contains the logo.

> >
> > I do wonder if the audio should be taken into account as well as the
> > video. If there were some way to measure the "density" of the audio
> > it would almost certainly be found to be higher on commercials.
> >
> > The is called "psycho-acoustic perceived loudness" by Orban and
> > others, and is why a commercial can sound louder than the program
> > even when your VU meter says it is the same level.
> >
> > Using obnoxious behavior by broadcasters and advertisers against them
> > is very satisfying. The annoying "logos" have turned out to be a
> > great way to detect commercial blocks. Sort of the "Judo" approach,
> > using your opponent's strength against him :-)
> >
>
> Granted it's been mentioned before (by me and another person I think),
> but couldn't commercial detection be enhanced using the VBI data and
> finding the program rating?  Judging from another thread today about
> 'parental blocks', commercials do not have ratings and could possibly
> open up another door for detection.  This is all assuming that the
> rating isn't carried over into the commercial from the program...

Any idea how to query this rating in myth on frame by frame basis?  If you can
figure out that part I can look into adding this method of detection to the
experimental flagger (if it looks useful).

Daniel




More information about the mythtv-users mailing list