Re: [mythtv-users] commercial flagging idea - commercial fingerprinting
I had been thinking about commercial fingerprinting for awhile, I can see a use for it to (perhaps) revoultionise TV advertising just as PVR's are revolutionising TV viewing. I believe it's possible for this to be turned into a mechanism that improves the effectivness of adverts, and allows for some form of targeted advertising. Let me give an example (assuming some form of perfect commercial detection/fingerprinting system) You watch numerous programs, without seeing the commercials, but Myth is quitely detecting commercials and indexing them, and maybe extracting them on the side. Then one day your car insurance is up for renewal, your myth box could now give you a list of all car insurance commercials. Additionally some form of rating system either from previous viewing habits, or external recomendations, could give the most appropriate companies first, e.g. best, cheapest, woman driver insurrance, over 50s, etc. If there was some form of recommendation system for commercials then it would be easy to find the funny, unusual or interesting commercials as well as the ones so bad you just have to watch them. In theory if this process worked, it would make the commercials more valuable (at least the good ones, and therefore improve the general quality of commercials?) and in that case broadcasters or advertisers should be more interested in letting us know when a commercial break happens, and the individual commercials themselves. I.E. Closed caption/VBI identification information as mentioned by Brad). ___ mythtv-users mailing list mythtv-users@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
Re: [mythtv-users] commercial flagging idea - commercial fingerprinting
Also, the fingerprint alogrithm could include time information. Obviously, shows (in America at least), have a set length between commercial breaks. Each episode might be different, but this is not determined by the individual broadcast stations, so someone watching an episode of Friends in Texas will have the same ammount of time between commercials as the person watching in New York. You've obviously never recorded late-night foodtv. I have recorded several episodes of Good Eats that are 10 minutes of show, 8 minutes of commercial, 12 minutes of show (or so). Granted, not different networks, but definitely different than other versions of the same episode. You also have to remember about all of those damn popup ad things that channels are adding in now, so you'd need to limit your fingerprint to only certain regions of the screen. You'd also have to take brightness/contrast/etc into account, since users will have those set differently for recording. Cool idea, though. -Chris ___ mythtv-users mailing list mythtv-users@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
Re: [mythtv-users] commercial flagging idea - commercial fingerprinting
On Apr 12, 2005 12:02 PM, Chris Petersen [EMAIL PROTECTED] wrote: You've obviously never recorded late-night foodtv. I have recorded several episodes of Good Eats that are 10 minutes of show, 8 minutes of commercial, 12 minutes of show (or so). Granted, not different networks, but definitely different than other versions of the same episode. This is true. I record Good Eats anytime it's on... I have over 100 episodes mythed =). I know that sometimes late at night they sometimes have no commercials. Still, that doesn't mean that the same philosophy wouldn't. Lets say the same episode ran with commercials, then a few months later, ran commercial free. The fingerprint would identify pre-commercial footage and post-commercial footage. If the commercial doesn't exist, first off mythcommflag would not mark it as a commercial anyway. Second, it could take the additional fingerprint and use some logic code that says look, if there was a commercial break here based on the fingerprint, then how come it would be 0 seconds long? Ignore. Or something to that effect You also have to remember about all of those damn popup ad things that channels are adding in now, so you'd need to limit your fingerprint to only certain regions of the screen. I don't think that'd be as big as a concern as you think. I know exactly what you are saying, but also take in to account that audio would/could be used in the fingerprint too. Just as in musicbrainz, if I have a MP3 of a song that was recorded for a studio track... if I upload the finger print, it'll match extrememly well (99% lets say) to the studio track MP3, but it might match 60% to the accoustic version, and 80% to the remix, etc... Either way, as long as you continue to think of it as a way to strengthen the existing mythcommflag job, and not as a replacement or the main factor then you should be fine at least in theory =D You'd also have to take brightness/contrast/etc into account, since users will have those set differently for recording. True, but still don't think it'll make a HUGE difference. Few percentage points here or there. The more people that upload their fingerprints, the smarter the overall fingerprint for the show gets. Cool idea, though. Thanks... now only if I could code =D ___ mythtv-users mailing list mythtv-users@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
Re: [mythtv-users] commercial flagging idea - commercial fingerprinting
Ant Daniel wrote: Then one day your car insurance is up for renewal, your myth box could now give you a list of all car insurance commercials. I, for one, have absolutely no need to resort to commercials when looking for products and services. I use a combination of research and referals to make purchasing decisions. If the developers have a laundry list of things to fix/add, adding code to make the commercials more accessible seems like a step backwards in my opinion. -- Meatwad ___ mythtv-users mailing list mythtv-users@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
Re: [mythtv-users] commercial flagging idea - commercial fingerprinting
On Apr 11, 2005 2:24 PM, *Matt* [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote: Hello All, =D Sorry this is long: For the short version, just read the first two paragraphs. I had an idea a few months ago, and just thought I'd throw it out there. I think that mythcommflag is getting better and better. However, sometimes there are times it'll goof up, and some shows are notroriously harder than others to flag correctly by their unique filmographic nature(Lost, Law Order, CSI, etc...). How about a method where users can validate mythcommflag's results and make that info available to the myth community to strengthen the accuracy of commercial flagging by using verified data. In a perfect world, all of the commercial breaks would all happen at the same time. Everywhere. In reality, they do not. Every master cable headend follows a local origination commercial insertion schedule sent daily by their traffic and billing department. LO comercials are notoriously inconsistent when they start and stop as they are cued by tones from the satellites and last until the TB schedule *thinks* they should end. Furthermore, LO commercials are stored and played back from either 1/2 or 3/4 VTR decks (old-n-busted) or by digital insertion systems (new-hotness). In either case, the tapes are remastered to the cableco's preferred format or encoded on commercial MPEG-2 gear and stored on VOD servers built for this purpose (e.g. SeaChange). This process would require every underpaid tech at every headend to get it right every single time. They don't. D*TV and Dish subscibers have it somewhat better but inconsistencies still occur. IIRC, this has been discussed here before in greater detail with a good number of pro's and con's. -- Meatwad [1] Who am I kidding? A perect world would have no commercials at all. ___ mythtv-users mailing list mythtv-users@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
Re: [mythtv-users] commercial flagging idea - commercial fingerprinting
Yup, you'll find threads (often including me) about man of these sorts of features. The long term version -- aggregating which parts of programs people watch at normal speed and which they don't -- provides the only undefeatable commercial detection algorithm, unless you are among the first to watch a show, but it's a long way off. Fingerprints based on various video and audio fingerprinting -- or even easier, on closed caption text when present -- is a lot easier.Some of these can actually be applied without collaboration, in that once you have identified a budweiser commercial by its closed caption text you can spot it again everywhere. (Of course not all commercials have captions.) A _really_ fancy fingerprinting system could actually fingerprint shows and notice when the same snippet of video appears in two different shows -- almost always a commercial, but not always always -- and then cleverly delete the redundant fragment and point to it. (Something myth can't yet do of course but may be planned for the future.) One reason that's interesting is that aside from spotting repeated commercials and repeated shows, it can also let you browse the commercials and start learning what they are. Ie. some people like to capture the movie trailers. Though web download is often a better idea there. ___ mythtv-users mailing list mythtv-users@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
