Talk:Video search engine

Finding video on the web used to be easy. You just went t stp gender based violenceo YouTube and hoped you could find what you love. Now there are hundreds of more choices with new niche content sites coming out just about every day. How does one find what they want to watch? A search on YouTube has become overwhelming and you simply cannot sift through hundreds of pages.

A new breed of video search engines (http: //video.2try4.com/] video.2try4], [[SearchForVideo, Pixsy, FLURL, etc.), has come out allowing people to search many sites. This helps expand a search, but it still leaves out a lot of material on smaller sites that don't have XML interfaces to the larger search engines. You are also faced with the problem of getting swamped with poor quality non-relevent videos. Another group of sites, like Videosift filter videos by a community of users. Other sites, like OVGuide, provide a user interface into many video content providers and search engines, which helps one find more content, especially from smaller sources. It also allows you to choose which channel to search to help narrow down your results.

Perhaps this is where the future lies in video search, since different sources will be best to give you proper results. For example, you might want to search a cooking site for professional cooking demos. The same results might be available on YouTube, but since the videos on the cooking site are pre-screened by human editors, you have a much better chance of actually finding what you need.

68.171.47.57 18:15, 22 November 2006 (UTC)

The multi web search engine 2try4 provides multi video search for most engines online. Enabling users to discover & multi play clips on video search engines, such as Youtube, yahoo, Google, grouper, flurl, searchforvideo, blinkx, aol, etc.

82.34.90.77 18:00, 05 February 2007 (UTC)


 * That's complete horse shit. All your site does is pop open a load of frames linked to other sites and tries to pass the search results off as its own. Quit spamming. -- Sakurambo 桜ん坊  09:03, 13 April 2007 (UTC)

Vdoogle
Vdoogle is a video search engine and does a fine job of finding the video the user is looking for. this is not advertising, but helping people get what they are looking for. they cant find a video so they come here, looking for a video search engine, and vdoogle does a good job finding that video so to not ad it to the list would be depriving people. Vdoogle is not an advertisement on wikipedia simply because vdoogle does not say, "Click here!!! Better than all others!!!!" we simply add the link, give a description. it is their choice to go or to not. — 67.182.0.3 (talk • contribs) has made few or no other edits outside this topic.

External links policy on Advertising and conflicts of interest states You should avoid linking to a website that you own, maintain or represent, as is the case in this situation. Unfortunately your conflict of interest editing involves contributing to Wikipedia in order to promote Vdoogle. Such a conflict is strongly discouraged. Your contributions to wikipedia under Nickboyett and IP's 67.182.4.58, 67.182.36.71, 67.182.36.111, 12.44.170.10, 67.181.201.132, 67.182.36.151, consist entirely of adding external links to Vdoogle on this and other articles and is considered WP:Spam. Looking through your multitude of IP's contributions as a whole, the all seem to be Vdoogle related only. Please do not continue adding links to your own websites to Wikipedia. It has become apparent that your account and IP's are only being used for spamming Wikipedia and for self-promotion.--Hu12 06:34, 2 July 2007 (UTC)

Limited article
This article appears to be limited to web based video content search. Video content search itself is far wider. Standard web based video content search is very much dependent on metadata tagging by the uploader and viewers of the video.

The discussion on the using the audio (speech to text) is another form of enabler. It had been funded in the States by DARPA among other research organizations.

True video content search on the hand, involves the ability to parse the video frame and extract relevant color, shape and texture data (eg indexing). There are different variations and approaches to the search aspect it would be based on the indexed information. 08:31, 11 October 2010 (UTC)~

Cognika
Added cognika as an entry. Full-disclosure: I do represent Cognika, however do not intend this to be advertising. Please advise if I have violated the ToU and how I might present Cognika as valid entry. Please do note, our work carries significant scientific weight, and we would like to inform Wikipedia users.

"It is generally acknowledged that visual search into video does not work well and that no company is using it publicly. Researchers at UC San Diego and Carnegie Mellon University have been working on the visual search problem for more than 15 years, and admitted at a "Future of Search" conference at UC Berkeley in the Spring of 2007 that it was years away from being viable even in simple search."

I would also challenge the above notion. Cognika's technology has been evaluated and is being by multiple US government, law enforcement agencies who are using it to visual search inside video. At the risk of being labeled as self-promoting I would like to rewrite the section about visual search and present novel approaches into visual search. --Cognika (talk) 00:59, 3 December 2012 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Video search engine. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive http://web.archive.org/web/20110708130446/http://escac.documentacioinformativa.com:80/sessio1/lectures/digition.pdf to http://escac.documentacioinformativa.com/sessio1/lectures/digition.pdf
 * Added tag to http://www.activamultimedia.com/am/v_portal/herramientas/generarpdf.php?te=26&idm=2

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 08:21, 21 July 2016 (UTC)

Speech recognition article
I would like to add to section Speech recognition a mention about project Spoketube, which does the same as Google Gaudi - uses voice recognition technology to locate the exact moment that one or more words have been spoken within an audio of YouTube video, allowing the user to go directly to the exact moment that the words were spoken. But Gaudi has not been available for public use for many years, unlike Spoketube. Wishes90 (talk) 00:39, 27 June 2020 (UTC)