WO2008126050A2

WO2008126050A2 - Embedding tv-anytime data in content

Info

Publication number: WO2008126050A2
Application number: PCT/IB2008/051413
Authority: WO
Inventors: Peter S. May; Robin J. Blackwell
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2007-04-17
Filing date: 2008-04-14
Publication date: 2008-10-23
Also published as: WO2008126050A3

Abstract

A first content item is steganographically encoded with a TV-Anytime (TVA) Content Reference Identifier (CRID). Metadata can also be encoded. A steganographic decoder (16) extracts the embedded CRID from the first content item. The extracted CRID can be used to request metadata corresponding to the CRID or to request the CRID is resolved to a locator of content corresponding to the CRID. The content item can form part of a web page and the decoder (16) can form part of a plug-in (14) to a browser (12). The CRID can reference a second content item or the CRID can provide an association to different versions or formats of the first content item. This enables tunnelling of TVA data through existing channels.

Description

Embedding TV-anytime data in content

FIELD OF THE INVENTION

This invention relates to distributing TV-Anytime data.

BACKGROUND TO THE INVENTION TV-Anytime (TVA) is a standardised set of specifications (ETSI 102 822), established by the TV-Anytime Forum, for delivering multimedia content over distribution networks such as broadcast networks and online systems. TV-Anytime provides a structure for metadata, a query language for searching, transport mechanisms for delivering the metadata and a mechanism for resolving between the metadata and the location of the content. Full details can be found at http://www.tv-anytime.org/. One important aspect of TV-Anytime is the separation of an identifier for a particular item of content from the storage location of that item of content. Content is referenced by a Content Reference Identifier (CRID). A CRID can be used to search and retrieve metadata relating to the content and a CRID can be resolved to the location of the content itself. As an example, a CRID representing a particular episode of a television programme can be resolved to a date, time and channel when a particular broadcast TV channel is next scheduled to transmit that content, or the CRID can be resolved to a URL of a network server which can provide that content. The storage locations can vary over a period of time. Furthermore, CRIDs can reference other CRIDs, enabling CRIDs to represent a group of related contents, such as a series.

TV-Anytime also defines an Interchange Data Format (TVA specification ETSI 102 822-8 vl.1.1) which is a data format for transferring information from a non-TVA world, such as a standard PC and web browser, to a TVA world, such as a TVA Personal Digital Recorder (PDR). It focuses on the case of an EPG being present on a web page and an application using the web page to extract the EPG information, followed by transmitting this information in the TVA Interchange Data Format to a PDR. The PDR may then perform some action, such as recording the content for later viewing. One problem which has prevented TV-Anytime from being widely adopted is that it requires some effort by the web hosting company to include, in some way, the required TVA data e.g. including a button on the page which will download the TVA data. This has the drawback that it requires hosting companies to be TVA aware and to adapt their existing sites. This restriction greatly limits the likelihood of adoption.

It has been proposed to use digital watermarking techniques to store a code within a content item for the purposes of copy protection. US 2003/0103645 describes a system of this kind, and also describes using a watermark as a link to metadata stored elsewhere which can identify the owner or content. An article "Applications for data hiding", W. Bender et al, IBM Systems Journal, pages 547-568, Vol.39 Nos 3 & 4, 2000, describes use of data hiding techniques for digital images. This also concentrates on anti-counterfeiting applications and mentions embedding a Uniform Resource Locator (URL) or a filename where additional information is stored.

The present invention seeks to provide an alternative way of making TV- Anytime data, such as CRIDs, available.

SUMMARY OF THE INVENTION

A first aspect of the present invention provides a method of retrieving TV- Anytime data comprising steganographically decoding a first content item which has been steganographically encoded with a TV-Anytime Content Reference Identifier (CRID) to extract the encoded Content Reference Identifier. An advantage arising from this invention is that existing content hosting web sites do not need to be modified to carry the TV-Anytime (TVA) data, since the steganographic encoding hides the TVA data in a manner which is transparent to the site hosting the content item. In effect, this enables tunnelling of TVA data through existing channels. This allows improved interoperability between the TVA aware world and the non- TVA world.

Another advantage arising from the invention is that content can be shared within a local (e.g. home) network, or externally of the local network, using standard file transfer techniques. TV-Anytime methods can then be used to retrieve metadata corresponding to the Content Reference Identifier or to retrieve a locator of content corresponding to the Content Reference Identifier.

The decoding can be performed on a content item which is stored locally with respect to where the decoding occurs (e.g. an item stored on a PC) or the decoding can be performed on a content item which has been accessed via a network connection, such as a content item on a web page. The CRID can reference other versions or formats of the content item. This is especially useful as web content is created for viewing on a PC and does not always render well on Consumer Electronics (CE) products, which have different quality of service expectations and differing levels of codec support. By embedding a CRID in a first content item which references other formats of the content item, and subsequently extracting a CRID, a rendering device can acquire an instance of the content that is most suited to its capabilities without the site hosting the first content item having to be specially modified to support a set of links to other formats.

The method can be performed by a terminal which is capable of browsing for, or viewing, content such as a personal computer (PC), a portable computer, a mobile phone, a media player, a set-top box, a media server or a personal digital recorder (PDR). The functionality can be provided as a stand-alone application or as a plug-in to an existing application, such as a plug-in to a browser.

A further aspect of the present invention provides a method of distributing T V- Anytime data, comprising: embedding a TV- Anytime Content Reference Identifier (CRID) within a first content item using a steganographic encoding technique, the Content Reference Identifier providing a reference to at least one of: metadata and content.

The embedding of a CRID can be performed by an external network entity, such as a Content Provider or web hosting service, or by a local network entity, such as a media server within a home network.

An embodiment of the invention is described in the context of a content item forming part of a web page. However, the content item does not have to form part of a web page. The invention can be generally applied to content items per se, or to content items forming part of any kind of document (e.g. word processing file, spreadsheet, presentation). The content items can be a still image, audio content, video content, games or any other content (in a compressed or uncompressed format) and the content may be stored on any suitable storage device or media such as a memory device, hard disk, optical disc or network server. The encoding of the CRID can use any known steganographic (data hiding) technique, such as least-significant bit coding and perceptual masking.

The functionality described here can be implemented in software, hardware or a combination of these. The invention can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. Accordingly, another aspect of the invention provides software for implementing the method. The software may be stored on an electronic memory device, hard disk, optical disk or other machine-readable storage medium. The software may be delivered as a computer program product on a machine-readable carrier or it may be downloaded to a device via a network connection.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will be described, by way of example only, with reference to the accompanying drawings in which: Fig. 1 shows a system according to an embodiment of the invention;

Fig. 2 shows an encoder used at a content provider; Fig. 3 shows an embodiment of apparatus at a user premises; Fig. 4 shows signalling flows of a process in which content is created, made available and selected by a user; Fig. 5 shows a user interface of a browser;

Fig. 6 shows a way of sharing metadata within a home network using the invention.

DESCRIPTION OF PREFERRED EMBODIMENTS Figure 1 shows an overview of a system for implementing the invention. The main network entities are a content provider 40, a web hosting server 50 and a TV-Anytime metadata server 60. Content provider 40 produces content. The content can comprise anything a user would like to consume, such as images, audio content, video content (movies, TV programmes etc.), games or other forms of content. The content provider 40 also provides a service to distribute the content. Server 50 hosts a web site on which is stored content items having embedded TVA Information (CRIDs) embedded within them. Metadata Server 60 stores metadata which is associated with content. Network entities 40, 50, 60 are shown separately but a single entity may perform several functions, e.g. a content provider may also respond to requests for metadata, or a function may be split among several entities, e.g. a content creator creates content and a content service provider provides a service to distribute the content. Network entities 40, 50, 60 are connected with users via a network 25 such as the Internet. Apparatus 10, 20 at one user premises is shown in Figure 1.

Figure 2 schematically shows apparatus for encoding content with a Content Reference Identifier (CRID). This apparatus can be located at the content provider 40 or the web hosting server 50. Content 41 is applied to a steganographic encoder 43. A CRID for content is generated by the content provider 40 or is issued by another network entity. The CRID is also applied to the steganographic encoder 43. The steganographic encoder 43 uses a steganographic encoding technique to transparently embed the CRID 42 within the content 41. The encoder 43 can also access a store 45, and can use data within the store 45 to control the way in which the encoder operates. Data within store 45 may specify a steganographic encoding technique for each content type (e.g. technique A for still images, technique B for video, technique C for audio etc.) The encoded content 44 is output, for transfer to a hosting site. The apparatus can embed metadata 46 as well as just a CRID. This metadata 46 could, for example, describe the content which the CRID references, or the content within which the CRID is embedded, such as a title of the content. The metadata could add some semantic information such as "this is a trailer for a movie".

Figure 3 shows one example of apparatus 10, 20 at a user's premises. This comprises a personal computer (PC) 10 and a web browser 12 for browsing content and a TVA-enabled Personal Digital Recorder (PDR) 20. In a conventional manner, the PC executes Internet browser software 12 which allows a user to browse web pages. The browser 12 accesses servers on the Internet via a network interface 11. The browser can have various software plug-ins to decode particular types of content. In this embodiment of the invention, a TV-Anytime plug-in 14 is provided for decoding CRIDs embedded within content appearing on a web page. The plug-in 14 communicates with the browser via an Application Programming Interface (API) 13. The plug-in 14 performs steganographic decoding 16 using one or, preferably, a plurality of different decoding techniques. The plug- in 14 can also access a store 15, and can use data within the store 15 to control the way in which the plug-in operates. As described more fully below, store 15 can store a list of internet content types which the plug-in should decode, and can store a list of URLs of pages which the plug-in should decode. The TVA browser plug-in 14 communicates with a PDR 20 via an interface 18, which can use the TVA Interchange Data Format to send extracted CRIDs to the PDR. The browser plug-in 14 acts as a bridge between the non-TVA and TVA aware domains. The PDR 20 comprises a controller 21 and a high-capacity store 22 such as a hard disk for storing content. The PDR has a user interface 24 and rendering capabilities 25 to render stored content and output video and audio signals 26, 27. A TV-Anytime module 23 connects to the controller 21 and can communicate with network servers via a network interface 26. The TVA module 23 can issue requests to resolve CRIDs and, in reply, will receive a further CRID or a locator of content corresponding to the CRID. TVA module 23 can also retrieve metadata corresponding to CRIDs. Controller 21 can use the locator obtained by TVA module 23 to acquire content from a network server, using network interface 26, or to acquire content from a broadcast source using a conventional DVB interface (not shown).

In Figure 3 a user browses for content using PC 10 and content is acquired and recorded using PDR 20. This requires cooperation between the PC 10 and PDR 20. The link between PC 10 and PDR 20 can be established manually or automatically using a home networking technology such as UPnP. The use of a browser plug-in is one way of implementing the decoding functionality at the client side and has an advantage that the browser remains unchanged (apart from the plug-in) and therefore the user's browsing experience is unaffected. The web browsing does not need to take place on a separate device to the TVA-enabled device. A TVA-enabled device with browsing capability, such as a TVA-enabled Web TV, may be used to perform all of the functions of the apparatus shown in Figure 3.

Figure 4 shows an example of a process by which content is created, made available and selected by a user. Communication protocols are provided as examples only. The first sequence of steps relate to creating and publishing content. Firstly, the Content Provider 40 produces content (step 80). A CRID is created for the content. Metadata can also be created at this step, and is published to the Metadata Server 60 (step 81). The metadata is associated with the CRID. There are several categories of TV-Anytime Metadata. Content description metadata describes the content, such as title, synopsis, keywords, credits, reviews, purchase information. At step 82 the CRID is embedded within a content item using a steganographic encoding technique. At step 83 the content item, with the embedded CRID, is uploaded to the web hosting server 50. The uploading may be achieved by File Transfer Protocol (FTP) or any other suitable transfer protocol. The hosting server 50 makes the newly uploaded content available to users browsing the Internet.

The next sequence of steps relate to a user browsing and acquiring content. At step 84 a user browses the site maintained on server 50. The user navigates to a web page which includes the content item having the embedded CRID. Because the CRID has been embedded in a transparent manner (i.e. there is no obvious flag to indicate that the CRID is present), the browser plug-in needs to test for the presence of the embedded CRID. There are several possible ways of achieving this: (i) the browser plug-in may automatically attempt to decode an embedded CRID in any content item appearing on a page by using one or more steganographic decoding techniques.

(ii) the browser plug-in can be configured to only attempt to decode certain types of content. The browser plug-in maintains a list of the internet content types (MIME types) of the content items in which a CRID may be embedded. The browser notifies the plug-in of the MIME type of the items on the page (step 86). If the MIME type matches one of the stored list of MIME types, the plug-in begins to decode the content item to test for the presence of an embedded CRID. This has an advantage of preventing unnecessary decoding operations, thereby conserving processing resources.

(iii) as an addition, or as an alternative, to (ii) the browser plug-in can maintain a list of web sites or pages (e.g. in terms of URLs) that carry content items having embedded CRIDs. If the current web page matches one of the stored list of URLs, the plug-in begins to decode the content item to test for the presence of an embedded CRID. For many types of content embedded within a web page, the plug-in 14 can immediately begin to decode the content to find an embedded CRID. For types of content (e.g. video content which requires a user to select the video before it will begin playing) a user may need to select the content item (step 85) before the plug-in 14 can begin decoding the embedded CRID. However, it is possible to work around this limitation by embedding the CRID within the part of the content item which is presented to a user before selection, such as the thumbnail of a video clip.

As an optional step, when a CRID is extracted from a content item the user can be notified at this point that a CRID has been found and the user can be given the option of proceeding to use the CRID. Various options for this stage are described later. At this point, the browser plug-in can retrieve metadata from metadata server 60 or this can be left until a later step (step 89).

The plug-in may proceed directly to step 88, and package the CRID into an appropriate format for sending to the TV-Anytime enabled device (PDR). The CRID can be packaged using the TVA Interchange Data Format, as described in ETSI 102 822-8, and sent using SOAP over HTTP. Other options are possible, such as TVA remote programming or a proprietary mechanism. The TVA device may acknowledge receipt of the data. The PDR makes a query for metadata to the TVA Metadata Server 60 (step 89). The query contains the CRID which has just been received. When metadata is received, it can be presented to a user. Additionally, the TVA device makes a query to resolve the CRID into a locator of content (step 90). The CRID resolution query is submitted to a resolving authority which, in this example, is the Content Provider 40. The query can use SOAP as described in ETSI 102 822-6-1. Results of the CRID resolution are received (step 91) and may be presented to a user before acquiring one of the instances of the content from the Content Provider 40 (step 92) or the PDR may automatically proceed to acquire the content.

Finally, the PDR records the content (step 93) or schedules a recording for the content. Depending on the instance of the content that is selected, the content may be acquired from a network server via a streaming or file transfer protocol or from a broadcast delivery channel, such as a Digital Video Broadcasting (DVB) channel. The request to resolve the CRID at step 90 can make use of stored capabilities of the PDR or rendering device, such as display resolution (standard or high-definition), supported codecs etc. so that an instance of the content is selected which best matches the capabilities of the device. The selection of the most suitable instance of the content can be made by the PDR 20, or the PDR 20 can submit device capability data to a server (metadata server or resolution server) in advance using the TVA submit mechanism. The server, upon receiving a CRID from the PDR 20, selects the most suitable instance of the content and sends a locator to that instance of the content to PDR 20. This can occur automatically, or following user input to select the required instance of the content. Other user preferences can also be submitted in this way, such as preferred genres, parental ratings etc. of content. These user preferences can either be manually set or can be inferred from viewing habits.

Steps 89, 90 assume there is a bi-directional communication channel between the PDR 20 and TVA metadata server 60. TVA can be applied in systems where there is only a uni-directional communication channel between a provider of metadata and the PDR 20, such as a broadcast system. In the case of a uni-directional channel, the PDR locally caches metadata received from the broadcast channel for use in resolving queries.

In the above process, the CRID is embedded within a content item by a content provider 40, before uploading the content to a web hosting site 50. Alternatively, the CRID may be embedded within a content item by the hosting site as part of the registration of new content. This could include requesting information from the content provider 40 to describe the content. The hosting site can then convert and embed this information in TVA form into the content item as a CRID with, or without, accompanying metadata.

There are several different ways in which the method described above can be used: (i) a CRID is embedded within a first content item; the CRID relates to a second content item. As an example, a short trailer of an upcoming movie or TV programme is published on a website. The CRID for the upcoming content is embedded within the trailer. The system acquires the CRID from the trailer and is then able to use that CRID to record the content, when it becomes available, to the PDR. Similarly, adverts placed on websites are embedded with CRIDs which cause content to be downloaded to the PDR.

(ii) a CRID is embedded within a first content item; the CRID provides an association to different versions or formats of the first content item. As an example, a low- resolution version of a first video content item is published on a website. A CRID identifying the content is embedded within the content item. When the CRID is extracted and resolved the TVA device is able to locate other formats of the item, such as a high- definition version suitable for viewing on a high-end system or a version which is optimised for viewing on a portable device. The extracted CRID may also be resolved to locate other versions of the item, such as a family- friendly or adult version of the content. The resolution process can use TVA instance metadata IDs (IMI).

(iii) a CRID is embedded within a content item; the CRID provides an association to metadata describing the content item.

In each case, embedding the CRID within a content item allows any existing website to host the content item without having to adapt the website to carry the CRID. All of the steps of the process described above can occur in real-time, or near real-time. While a user browses content, an embedded CRID is extracted from content and used as a reference to acquire metadata and/or content based on the CRID. However, the steps of the process can occur at different times. For example, an offline system can decode any embedded CRIDs in real-time and store the CRIDs in a cache and, at a later time, use the CRIDs to obtain metadata and/or content. Alternatively, an offline system can store a copy of whatever content a user browses and all decoding to extract embedded CRIDs, and using the CRIDs to acquire metadata and/or content, can occur at a later time. A CRID can be created for future content which is not yet available, and embedded within a different content item. For example, a CRID may reference a movie which has not yet been released, and is embedded within a trailer for the movie. When PDR receives a CRID for content which is not yet available, the PDR can periodically check if it is available. As a further option, the plug-in could also perform web crawling on the user's behalf, searching web pages on the Internet, extracting CRIDs and using the CRIDs to download content to the PDR 20 that is recommended based on the user's profile. A TV-Anytime CRID can take the form of a group CRID. A group CRID can refer to a series of programmes (e.g. a soap opera) and can be resolved to a group of CRIDs each referring to a particular episode. A TV-Anytime CRID can take the form of a package CRID which references a package of content items, such as interactive content, games, alternative audio tracks, ambience etc. The metadata returned at step 91 can include TVA material fields which could point to a website, email address, item for purchase etc.

During the browsing process (steps 84-87) there are various ways in which decoded CRIDs can be managed and presented to a user. Figure 5 shows a graphical user interface window 100, such as the normal window of a web browser 12, and some examples of how a user can be notified of decoded CRIDs. The user interface may use one or more of these techniques. The browser plug-in 14 can maintain a list of any decoded CRIDs which have been found on web pages visited by the user. This is shown as a 'CRID history' item

101 in the main toolbar of the window 100 and a drop-down menu 102. Each entry in menu

102 comprises a CRID 103 and can also include a thumbnail 104 to remind the user of the item the CRID was embedded within. A user may take an action for an entry in the menu, such as to use the CRID, which would cause the CRID to be transferred to PDR 20 and metadata to be retrieved or the CRID to be resolved. A pop-up window 105 can appear next to, or on top of, an item 106 for which an embedded CRID has been decoded. A user can then decide whether to use the CRID (e.g. triggering the transfer of the CRID to the PDR 20 and an instruction to record content corresponding to the CRID), store the CRID in menu 102 or simply ignore the CRID. The various task options can be presented in response to a user right-clicking on the pop-up window 105. As a further example, the cursor 107 may change from the regular cursor shape to a different cursor shape indicating that the item currently beneath the cursor has a CRID embedded within it. The user may also set-up particular behaviour for processing decoded CRIDs, such as "only use CRIDs found on web page with this URL", "record content for any decoded CRIDs on device X". As a further option a media player or media player plug-in, such as a Flash player, which is used to render browsed content items can be modified to support the functions described here of displaying CRIDs and giving the user an option to record content when an embedded CRID is found. If some embedded metadata accompanies the CRID, this can be presented with the CRID in any of the options described here, or the plug-in can include a TVA function which allows it to query a metadata server 60 to obtain metadata which can be presented alongside the CRID. If the plug-in allows a user to select whether to acquire/record content then preferably the plug-in should be aware of the capabilities of the PDR which will record the content or the capabilities a display device which will render the content.

The process of encoding a content item with a CRID can use conventional steganography techniques, such as those known to embed copy protection information within content. The following describes a known example of encoding data within a still image, which can be used to embed a CRID within a still image. Each pixel of a full colour image is typically represented by 24 bits (ignoring transparency) - 1 byte for each of the RGB (red, green, blue) colour components. A single byte represents a value (between 0 and 255 inclusively) indicating the intensity of that colour component. The least-significant bit (LSB) of a byte can be modified (e.g. 1 to 0, or vice versa) without causing a perceptible difference to the naked eye, and so this bit can be used to encode information. On this basis, an 8-bit ASCII character can be encoded in 3 pixels (3 pixels encodes 9 bits of information). To encode a full CRID, multiple groups of 3 pixels are used in the above way. The modified pixels can be located contiguously within the image or in positions which are known both to the encoder and decoder. The CRID can be embedded multiple times within an image, at different spatial and/or temporal parts of a content item (e.g. different frames of a video sequence, different portions of an audio item. One disadvantage of this approach is that it suffers from robustness issues. The embedded information can be removed simply be setting all LSBs to zero, without dramatically affecting the quality of the image. Also, the information can be affected if the image is modified (scaling, rotation, cropping, etc). There are a wealth of alternative techniques which all have various advantages, such as Discrete- Cosine Transform, Discrete-Fourier Transform, and perceptual masking techniques. Various suitable techniques are described in the article "A Review of Data Hiding in Digital Images", Eugene T. Lin and Edward J. DeIp, Video and Image Processing Laboratory (VIPER), School of Electrical and Computer Engineering, Purdue University, available at: ftp://skynet.ecn.purdue.edu/pub/dist/delp/pics99-stego/paper.pdf.

To ensure that the TVA information embedded within a content item is not corrupted, standard steganography techniques can be used such as checksums etc. In addition, the CRID itself can be digitally signed as described in International Patent Application WO 04/036870.

Figure 6 shows a further embodiment of the invention. A home network comprises a plurality of devices 91, 92, 93 which are interconnected 90. A media server has a store 94 for storing content items 98 Content items 98 are steganographically embedded with a CRID in the manner previously described. Store 94 also stores metadata corresponding to the content items 98. Each piece of metadata is associated with a CRID, which in turn is associated with a particular content item. Other devices 91, 92 in the network can request content items 98 from the media server 93. Figure 6 shows a content item being delivered 95 to device 91 in response to a request. Device 91 has a decoder to steganographically decode the received content item and extract the CRID which server 93 has embedded within the item. Device can subsequently issue a request 96 for metadata and receive metadata 97. The association between a CRID, content item and metadata may already have been created by an external server, with the media server 93 storing content 98 and metadata with this pre-existing association. Alternatively, media server 93 can create a CRID to associate content with metadata (or other content). In this option, the media server has the functionality shown in Figure 2 and acts in the same way as an external content provider 40. One application for this is as follows. A digital camera is connected to a PC and a user uploads images to the PC in the normal way. The user enters metadata for each image (e.g. title, place) and then the PC generates a unique CRID for the image, and embeds this in the photo using a stenographic encoder. Metadata could alternatively be entered on the camera and transferred to the PC with the uploaded images. The PC could extract certain data (e.g. the date) from the camera image format. Each digital image in a library of digital images can have a unique CRID embedded within it. The CRID is linked to metadata describing that image, such as date, place, comments or other content such as audio acquired at the time of taking the image. The PC supports a TVA Server with the metadata. Any form of standard photo sharing server application (e.g. UPnP based) could be used to distribute the image within the home network, or the image can be uploaded to a photo storage website externally of the home network, with the CRID embedded within the image. Any receiving device can decode the CRID and then query the TVA server hosted by the PC. Content can be shipped between devices 91, 92, 93 without losing the CRID. The home network could exist on top of the UPnP AV architecture. A new communication protocol is not required and the file formats would not need to be changed to transport metadata with a content item. The receiving end device 91, 92 needs to support the particular steganographic technique used by server 93 but any intermediary devices within cloud 90 do not need to be capable of steganographic decoding since the CRID is tunnelled across network cloud 90, within the content item. This would also ensure that the metadata provided is up to date. This method is particularly advantageous as transfer of metadata is not standardised, and it is possible for metadata to become lost if it is sent using a separate transfer mechanism and tied to a content item in some other way. Digital watermarking is already used to identify copyright infringement. The addition of a CRID to the watermark would provide all the benefits of watermarking plus some extended features. A combined system would be able to use the rights management techniques of TVA to ensure that content is not simply copied from the web. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practising the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:

1. A method of retrieving TV-Anytime data comprising: steganographically decoding (16) a first content item which has been steganographically encoded with a TV-Anytime Content Reference Identifier (CRID) to extract the encoded Content Reference Identifier.

2. A method according to claim 1 further comprising using the extracted Content Reference Identifier by requesting metadata corresponding to the Content Reference Identifier (89) or requesting the Content Reference Identifier is resolved (90) to a locator of content corresponding to the Content Reference Identifier.

3. A method according to claim 2 wherein the first content item is different to the content corresponding to the Content Reference Identifier.

4. A method according to claim 2 wherein the Content Reference Identifier associates the first content item with at least one different format or version of the first content item.

5. A method according to claim 4 further comprising storing capabilities of a TV-Anytime device (20) and using the stored capabilities when requesting the Content Reference Identifier is resolved to a locator of content or selecting an instance of the content to acquire.

6. A method according to any one of claims 2 to 5, further comprising storing a user preference and using the stored preference when requesting the Content Reference Identifier is resolved to a locator of content or selecting an instance of the content to acquire.

7. A method according to any one of claims 2 to 6, further comprising prompting a user about the presence of an extracted Content Reference Identifier before using the extracted Content Reference Identifier.

8. A method according to any one of the preceding claims further comprising storing a list of content types for which the steganographic decoding should occur and automatically decoding a content item when the content type matches one of the stored list of content types.

9. A method according to claim 8 wherein there is a plurality of different steganographic decoding techniques and the stored list of content types specifies a steganographic decoding technique for use with each content type.

10. A method according to any one of the preceding claims wherein the first content item has been steganographically encoded with metadata (46), the method further comprising steganographically decoding (16) the first content item to extract the metadata.

11. A method according to any one of the preceding claims wherein the content item which has been steganographically encoded with a TV-Anytime Content Reference Identifier forms part of a web page.

12. A method according to claim 11 wherein the decoding is performed by a browser plug-in (14), the browser plug-in (14) having an interface for communicating the extracted Content Reference Identifier to a TV-Anytime enabled device (20).

13. A method according to claim 2 further comprising acquiring (92) an instance of the content from the locator and recording (93) the acquired content.

14. A method of distributing TV-Anytime data, comprising: embedding a TV-Anytime Content Reference Identifier (CRID) within a first content item (41) using a steganographic encoding technique (43), the Content Reference Identifier providing a reference to at least one of: metadata and content.

15. A method according to claim 14 further comprising embedding metadata (46) within the first content item (41) using a steganographic encoding technique (43).

16. A method according to claim 14 or 15 wherein there is a plurality of different steganographic encoding techniques and the step of embedding a Content Reference Identifier comprises selecting one of the steganographic encoding techniques based on a type of the content item.

17. A method according to any one of claims 14 to 16 wherein the first content item is different to the content referenced by the Content Reference Identifier.

18. A method according to any one of claims 14 to 16 wherein the first content item is a different format or version of the content referenced by the Content Reference

Identifier.

19. A method according to any one of claims 14 to 18 further comprising embedding the first content item within a web page.

20. A method according to any one of claims 14 to 19 further comprising creating the Content Reference Identifier for embedding within the first content item.

21. A signal resulting from the method of any one of claims 14 to 20.

22. Software for causing a processor to perform the method according to any one of claims 1 to 13.

23. Software according to claim 22 in the form of a plug-in (14) for a web browser (12).

24. Software for causing a processor to perform the method according to any one of claims 14 to 20.

25. Apparatus comprising a processor arranged to perform the method according to any one of claims 1 to 20.