Automatic content recognition

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

Automatic content recognition[1] (ACR) is a technology to identify content played on a media device or present within a media file. Devices implementing ACR can allow the device or its manufacturer to collect content consumption information automatically "at the screen level" without any user-based input or search effort. This information may be collected for purposes such as personalized advertising, content recommendation, or for sale to customer data aggregators, and for other applications.[1][2]

How it works[edit]

To start the recognition, a short media clip (audio, video, or both) is selected from within a media file or captured as displayed on a device such as a smart TV. Using techniques such as fingerprinting and watermarking, the selected content is compared by the ACR software with a database of known recorded works.[3] If the fingerprint of the media clip finds a match, the ACR software returns the corresponding metadata regarding the media as well as other associated or recommended content back to the client application for display to the user or for collection by the device manufacturer or a data aggregator.[4]

Fingerprints and watermarking[edit]

Two leading methodologies for audio-based ACR are acoustic fingerprinting and watermarking. Similarly, video fingerprinting is used to facilitate ACR for visual media.

Acoustic fingerprinting generates unique fingerprints from the audio content itself. Fingerprinting techniques are agnostic to content format, codec, bit rate and compression techniques.[5] This makes it possible to employ across varying networks and channels[clarification needed] and is widely used for interactive TV, second screen application, and content monitoring sectors.[6][7] Popular apps like Shazam, YouTube, Facebook,[8] Thetake, WeChat and Weibo reportedly use audio fingerprinting methodology to recognize content played from a TV to trigger additional features like votes, lotteries, topics or purchases.[citation needed]

In contrast to fingerprinting, digital watermarking require the inclusion of digital "tags"[further explanation needed] embedded within the digital content stream prior to distribution. For example, a broadcast encoder might insert a watermark every few seconds that could be used to identify to broadcast channel, program id, and time stamp. This watermark is normally inaudible or invisible to the users. Display devices like phones or tablets however can read the watermarks to identify the content it is playing.[9] Watermarking technology is also utilized in the media protection field to help identify where illegal copies originate.[10]

History[edit]

In 2011, ACR technology was applied to TV content by the Shazam service, which captured the attention of the television industry. Shazam was previously a music recognition service which recognized music from sound recordings. By utilizing its own fingerprint technology to identify live channels and videos, Shazam extended their business to television programming. Also in 2011, Samba TV, at the time known as Flingo,[11] introduced its patented video ACR technology, which uses video fingerprinting to identify on-screen content and power cross-screen interactive TV apps on Smart TVs.[12] In 2012, satellite communications provider DIRECTV partnered with TV loyalty vendor Viggle to provide an interactive viewing experience on the second screen. In 2013, LG partnered with Cognitive Networks (later purchased by Vizio and renamed Inscape), an ACR vendor, to provide ACR driven interaction.[13] In 2015, ACR technology spread to even more applications and smart TVs. Social applications and TV manufacturers like Facebook, Twitter, Google, WeChat, Weibo, LG, Samsung, and Vizio TV have used ACR technology either developed by themselves or integrated from third party ACR providers.[citation needed] In 2016, additional applications and mobile OS embedded with automatic content recognition services were available including Peach, Omusic and Mi OS.[14][15][16]

Applications[edit]

Advertising and customer data collection[edit]

Data collected on the media consumption habits of customers is of enormous value to device manufacturers, advertisers, and data aggregation companies. ACR technology helps these companies surveil the interests of customers and collect data so that they can be more precisely targeted with personalized marketing and advertising campaigns. This data is extremely valuable-- it was reported in Nov 2021 that smart television manufacturer Vizio is more profitable from the sale of their customers' data than from the televisions they sold.[17]

Broadcast monitoring[edit]

For advertisers and content owners, it is vital to know when and where their content has been played. Traditionally agencies or advertisers have to manually audit the presentation. At scale it only can be checked through a statistical sampling method. ACR technology enables automatic near real-time monitoring of the content customers play on their TVs. Information like the time of play, duration, frequency, can be achieved without any manual effort.[18][19]

Audience measurement[edit]

Real-time audience measurement metrics are now achievable by applying ACR technology into smart TVs, set top boxes and mobile devices such as smart phones and tablets. This measurement data is highly essential to quantify audience consumption to set advertising pricing policies.

Content identification[edit]

ACR technology helps audiences retrieve information about the content they watched or listened to.[20] The identified video and music content can be linked to internet content providers for on-demand viewing, third parties for additional background information, or complementary media.

Content enhancement[edit]

Because devices can be "aware" of content being watched or listened to, second screen devices can feed users complementary content beyond what is presented on the primary viewing screen. ACR technology can not only identify the content, but also it can identify the precise location within the content. Thus, additional information can be presented to the user. ACR can enable a variety of interactive features such as polls, coupons, lottery or purchase of goods based on timestamp.[21]

Privacy Concerns[edit]

Organizations ranging from consumer rights advocates Electronic Frontier Foundation to tech web sites such as PCMag have expressed serious objections to the collection of user viewing consumption habits by their devices on privacy grounds.[22][23] However there is an option in almost every set to disable this feature.[citation needed]

Technology providers[edit]

ACR service providers include ACRCloud, Red Bee Media, Digimarc, Gracenote, Kantar Media, Inscape Data Services, Zapr Media Labs, Shazam and Samba TV.

See also[edit]

References[edit]

  1. ^ a b "Automatic Content Recognition (ACR)". Gartner. Retrieved 21 June 2020.
  2. ^ Lawler, Richard (2021-11-10). "Vizio's profit on ads, subscriptions, and data is double the money it makes selling TVs". The Verge. Retrieved 2021-11-22.
  3. ^ "ACR(Automatic Content Recognition)". Archived from the original on 28 February 2017. Retrieved 27 February 2017.
  4. ^ "Automated content recognition creating content aware ecosystems" (PDF). Civolution. Civolution. Archived from the original (PDF) on 23 September 2015. Retrieved 24 June 2015.
  5. ^ "Panako: a scalable acoustic fingerprinting system handling time-scale and pitch modification". Universiteit Gent. Retrieved 27 February 2017.
  6. ^ Main, Sami. "Nielsen Is Bringing Real-Time Interactive Ads to Smart TVs to Keep Streaming Audiences Engaged". Adweek. Retrieved 2018-01-11.
  7. ^ Brink, Kyle. "A Primer on Automated Content". Viggle. Viggle. Archived from the original on 2015-06-24. Retrieved 22 June 2015.
  8. ^ "Facebook Automatic Content Recognition". Starcom MediaVest Group. SMG. Archived from the original on 6 July 2015. Retrieved 6 July 2015.
  9. ^ Brink, Kyle. "SVP of Product Development". A Primer on Automated Content Recognition. Viggle. Retrieved 22 June 2015.
  10. ^ Solana, Anna. "How these hidden video watermarks can help spot piracy, doctored images | ZDNet". ZDNet. Retrieved 2018-01-11.
  11. ^ Baumgartner, Jeff (2013-09-24). "Flingo Rebrands as Samba TV". Multichannel News. Retrieved 2021-10-05.
  12. ^ Swedlow, Tracy (July 7, 2011). "Interactive TV News Round-Up (II): Flingo, Hulu, ITU".
  13. ^ "LG partners with Cognitive Networks to make Smart TVs smarter and more interactive". engadget. Retrieved 23 August 2016.
  14. ^ "ACRCloud Powers Song Recognition For Hottest New Social Network, Peach". Music Industry News Network. Music Industry News Network. Archived from the original on 8 March 2016. Retrieved 3 March 2016.
  15. ^ Victoria, Ho. "Xiaomi will help you name that song you can't stop humming". Mashable. Mashable. Retrieved 3 March 2016.
  16. ^ "ACRCloud Powers The Launch Of Taiwan's First Music/Humming Recognition Service For Omusic". Music Industry News Network. Archived from the original on 8 March 2016. Retrieved 3 March 2016.
  17. ^ Dunn, Thom (2021-11-18). "TV manufacturer Vizio makes more money selling data than TVs". Boing Boing. Retrieved 2021-11-22.
  18. ^ "Automated Content Recognition creating content aware ecosystems" (PDF). csimagazine. Civolution. Retrieved 24 June 2015.
  19. ^ "Roku Privacy Policy (Section I-B-4 and IV-E)". roku.com. Roku. Retrieved 30 October 2017.
  20. ^ Weiss, Tom (January 23, 2018). "Tom Weiss: Breaking the barriers to addressable advertising in Europe". Broadband TV News. Retrieved 30 August 2018.
  21. ^ Wolf, Michael. "Three Ways Automatic Content Recognition Will Change TV". Forbes. Retrieved 20 June 2015.
  22. ^ "Samsung, LG, and Vizio smart TVs are recording—and sharing data about—everything you watch Consumer Reports investigates the information brokers who want to turn your viewing habits into cash". Consumer Reports. Retrieved 27 February 2017.
  23. ^ "How to Stop Smart TVs From Snooping on You". PCMAG. Retrieved 2021-11-22.