I believe a time check was in the code, something like what you are
describing but I am not 100% sure. I reworked the code many times but we are
relying on MPlayer mostly, so if it's garbage in, it would be garbage out. \
This actually seems like a great MPlayer add on, if anyone here is familure
2009/4/21 Justin Wetherell :
> I wrote the commercial detection code for Freevo a long time ago. While I
> agree that it isn't accurate or great, it's the _only_ option available. And
> unfortunetly the MythTV developers aren't very code friendly.
>
> The three key indicators that MythTV talks abou
I wrote the commercial detection code for Freevo a long time ago. While I
agree that it isn't accurate or great, it's the _only_ option available. And
unfortunetly the MythTV developers aren't very code friendly.
The three key indicators that MythTV talks about are wonderful ideas. But at
the time
Bill Bierman wrote:
> The article I read that motivated me to give mythtv a try was this:
>
> http://www.mythpvr.com/mythtv/how-commercial-flagging-works.html
>
> As an aside, with both freevo and mythtv, I find getting sound to work
> well (both in live viewing and in recording) to be inordinate
b...@thebiermans.org said:
> As an aside, with both freevo and mythtv, I find getting sound to work well
> (both in live viewing and in recording) to be inordinately difficult. I
> consider myself to be extremely competent when it comes to computers, and I
> wasn't able to get sound working in my
The article I read that motivated me to give mythtv a try was this:
http://www.mythpvr.com/mythtv/how-commercial-flagging-works.html
As an aside, with both freevo and mythtv, I find getting sound to work
well (both in live viewing and in recording) to be inordinately
difficult. I consider myself
On Thu, April 16, 2009 9:19 pm, Bill Bierman wrote:
> Actually I didn't even mean to send those attachments. That was just
> my attempt at making the commercial detection more readable (and
> hopefully a bit more accurate). It was in the course of developing
> this patch that I realized the funda
Actually I didn't even mean to send those attachments. That was just
my attempt at making the commercial detection more readable (and
hopefully a bit more accurate). It was in the course of developing
this patch that I realized the fundamental flaw in the approach. When
I first drafted the messa
Bill Bierman wrote:
> I thought maybe I should post my final conclusions for people who may
> run into this issue in the future. Commercial detection as
> implemented by Freevo is unreliable. Period. It searches for 'black
> frames' it assumes are the beginning and end of commercials. It
> cann
I didn't want to get into them because it would require mentioning the
'competition' on the list -- but the other methods I have read about
have all come from the mythtv documentation.
On Wed, Apr 15, 2009 at 4:25 AM, Evan Hisey wrote:
> 2009/4/14 Bill Bierman :
>> I thought maybe I should post m
2009/4/14 Bill Bierman :
> I thought maybe I should post my final conclusions for people who may
> run into this issue in the future. Commercial detection as
> implemented by Freevo is unreliable. Period. It searches for 'black
> frames' it assumes are the beginning and end of commercials. It
>
I thought maybe I should post my final conclusions for people who may
run into this issue in the future. Commercial detection as
implemented by Freevo is unreliable. Period. It searches for 'black
frames' it assumes are the beginning and end of commercials. It
cannot account for the fact that t
I'm using tvtime, though I don't remember why =)
I've inserted a few debug statements in the commdetectcore.py code
which parses the mencoder output. Hopefully this will shed some light
on the situation.
Incidentally, the mencoder vf_blackframes output is hideous. It's
neither human-friendly no
On Tue, Mar 17, 2009 at 2:39 AM, Bill Bierman wrote:
> I'm actually using tvtime, but I have manually ran mplayer using the
> EDL files generated by Freevo, and it still doesn't work.
>
> I assume the problem is with the mencoder execution, or possibly the
> parsing of its output, as the EDL file
I'm actually using tvtime, but I have manually ran mplayer using the
EDL files generated by Freevo, and it still doesn't work.
I assume the problem is with the mencoder execution, or possibly the
parsing of its output, as the EDL file generated does not even seem
sane.
"3544 2603 0" is the single
On Sat, Mar 14, 2009 at 5:42 PM, Bill Bierman wrote:
> I have tried viewing the AVI file from Freevo, and there are still
> commercials in the recordings. I suppose this makes sense, though,
> since apparantely no commercials are being detected.
Is freevo using mplayer or xine to view the avi? X
I have tried viewing the AVI file from Freevo, and there are still
commercials in the recordings. I suppose this makes sense, though,
since apparantely no commercials are being detected.
I noticed that mencoder takes a parameter for 'blackness' in the black
frames filter which defaults to I belie
On Fri, Mar 13, 2009 at 12:40 PM, Bill Bierman wrote:
> Greetings. I am running Freevo 1.8.3 on Ubuntu. It is a stock
> installation with the exception of changing the order of signals sent
> to kill the webserver as mentioned by Duncan Webb in a previous post
> to this list.
>
> I am running th
18 matches
Mail list logo
