Event discovery and detection in interactive content

Abstract

A system and method for discovering and detecting events in interactive digital content is provided. Multiple game titles can be hosted on a game network platform, and game play data in sessions involving the multiple game titles can be tracked in real time. Each session can be associated with a stream. One or more event criteria can be stored for each of a plurality of predefined events of interest. When the game play data tracked for a session meets the stored criteria, an event of interest can be identified. A list of available streams identified as including the identified event can be classified, and a subset of the available streams can be identified as being at the top of the list. A notification can be provided to a viewer device, the notification including a link to one of the streams in the subset. The criteria for the event of interest can also be updated based on subsequent game data and feedback (e.g., use of links).

Description

Background Art

[0001] 1. Field of the Invention

[0002] The present invention relates generally to interactive content and more particularly to discovery and detection of events that occur during a session of interactive content.

[0003] 2. Related technical description

[0004] Currently, a wide variety of interactive digital content is available for viewing via online platforms such as Twitch. In the community of such online content consumers, skilled or otherwise famous players can become influencers and attract a large number of followers. Similarly, the popularity of various online digital games has also increased significantly, attracting an increasing number of participants and viewers.

[0005] However, viewers may face difficulties in discovering or identifying content of interest. Viewers of games can utilize information streams from various online media platforms. Such platforms may publish numerous games from numerous players (including those not participating in the tournament), however, making it difficult or at least inconvenient for viewers to identify games that may be of interest. This inconvenience carries the risk of losing interested audiences.

[0006] The ability to attract, retain, and encourage repeat viewers is particularly important for professional and semi-professional players who may receive sponsorship for their games. In addition to players who may receive sponsorship, commentators and other individuals who may seek to provide content related to tournaments may also face various difficulties in obtaining tournament data.

[0007] Particularly difficult for viewers is to identify events of interest in near real-time, so as to be able to access the associated stream. For example, a particular gaming session may involve a skilled player who is close to breaking a historical record. Viewers may wish to see the actual record being broken in as close to real-time as possible. However, viewers currently have no way of knowing when such an event will occur unless they are aware of such a record, are already watching, or are notified by a friend who is currently watching. However, verbal narration may not be particularly efficient or effective in notifying the intended viewer, who may not be able to access the stream in real-time. In some cases, such delays may mean that viewers may be spoiled for the outcome, or may be forced to wait for the stream to be archived and made accessible.

[0008] Therefore, there is a need in the art for improved systems and methods for discovering and detecting events that occur during playback of interactive digital content. Summary of the invention

[0009] Embodiments of the present invention enable discovery and detection of events occurring during the playback of interactive digital content. Multiple game titles can be hosted on a game network platform, and game play data in sessions involving the multiple game titles can be tracked in real time. Each session can be associated with a stream. One or more event criteria can be stored for each of a plurality of predefined events of interest. When the game play data tracked for a session meets the stored criteria, an event of interest can be identified. A list of available streams identified as including the identified event can be classified, and a subset of the available streams can be identified as being at the top of the list. A notification can be provided to a viewer device, the notification including a link to one of the streams in the subset. The criteria for the event of interest can also be updated based on subsequent game data and feedback (e.g., the use of links).

[0010] Various embodiments may include a system for discovering and detecting events occurring during playback of interactive digital content. Such a system may include at least one gaming network server that hosts a plurality of gaming titles and tracks real-time game play data in sessions involving the plurality of gaming titles, wherein each session is associated with a stream. The system may also include at least one analytics server that stores one or more event criteria for each of a plurality of predefined events of interest; identifies an event when the game play data tracked for a session meets the stored criteria; sorts a list of available streams identified as including the identified event; identifies a subset of the available streams as being at the top of the list; and provides a notification to a viewer device including a link to one of the streams in the subset.

[0011] Further embodiments may include a method for discovering and detecting events occurring during playback of interactive digital content. Such a method may include: hosting a plurality of game titles; tracking real-time game play data in sessions involving the plurality of game titles, wherein each session is associated with a stream; storing one or more event criteria for each of a plurality of predefined events of interest; identifying an event when the game play data tracked for a session meets the stored criteria; sorting a list of available streams identified as including the identified event; identifying a subset of the available streams as being at the top of the list; and providing a notification to a viewer device including a link to one of the streams in the subset.

[0012] Further embodiments may include a non-transitory computer-readable storage medium having embodied thereon a program executable to perform such methods as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is an exemplary network environment in which a system for discovering and detecting events occurring during playback of interactive digital content may be implemented.

[0014] Figure 2 is a block diagram of an exemplary stream selection engine that may be used in a system for discovering and detecting events that occur during playback of interactive digital content.

[0015] Figure 3 is a flow chart illustrating an exemplary method for discovering and detecting events during playback of interactive digital content.

[0016] Figure 4 An example of an electronic system that can be used to implement some aspects of the present technology is shown. DETAILED DESCRIPTION

[0017] Embodiments of the present invention enable discovery and detection of events occurring during the playback of interactive digital content. Multiple game titles can be hosted on a game network platform, and game play data in sessions involving the multiple game titles can be tracked in real time. Each session can be associated with a stream. One or more event criteria can be stored for each of a plurality of predefined events of interest. When the game play data tracked for a session meets the stored criteria, an event of interest can be identified. A list of available streams identified as including the identified event can be classified, and a subset of the available streams can be identified as being at the top of the list. A notification can be provided to a viewer device, the notification including a link to one of the streams in the subset. The criteria for the event of interest can also be updated based on subsequent game data and feedback (e.g., the use of links).

[0018] Figure 1 An exemplary network environment 100 in which a system for discovering and detecting events occurring during playback of interactive digital content may be implemented is shown. The network environment 100 includes a public network 102, which may include one or more private networks, such as a local area network (LAN), a wide area network (WAN), or a network of public / private networks, such as the Internet. The Internet is a widespread network of interconnected computers and servers that allows Internet Protocol (IP) data to be transmitted and exchanged between user devices connected through a network service provider. Examples of network service providers are a public switched telephone network, a cable service provider, a digital subscriber line (DSL) service provider, or a satellite service provider. The public network 102 allows communication between various components in the network environment 100. The public network 102 is communicatively coupled to a gaming network 104, which may also provide a network of computing resources configured to implement an event detection system 112.

[0019] Spectators may include players 108-110 who may access the gaming network 104 and event detection system 112 using their respective client devices (e.g., 108A-110A). Although the client devices 108A-110A are depicted as gaming console devices, it should be understood that the client devices may include various types of processor-based systems, including, but not limited to, personal computing devices, smart phones, tablet computers, other portable gaming devices, and the like. Such client devices 108A-110A may also be configured to access data from other storage media, such as memory cards or disk drives, which may be suitable for the case of downloaded services. The client devices 108A-110A may include standard hardware computing components, such as network and media interfaces, non-transitory computer-readable storage devices (memory), and a processor for executing instructions that may be stored in the memory. Reference is made below to Figure 4 Examples of hardware systems that may be used to implement the client device are discussed in more detail.

[0020] The audience may also include developers or other parties (e.g., 105-107), who may also access the gaming network 104 and event detection system 112 through their respective computing systems (e.g., 105A-107A). The developer computing systems 105A-107A may include any type of server or other computing device known in the art, including standard hardware computing components, such as network and media interfaces, non-transitory computer-readable storage devices (memory), and processors for executing instructions that may be stored in the memory or accessing information that may be stored in the memory. The functionality of multiple servers may be integrated into a single server. Any of the above servers (or integrated servers) may have certain client-side, cache, or proxy server features. These features may depend on the specific network placement of the server or certain configurations of the server. It should be understood that a greater (or lesser) number of players and / or developers may participate in the environment 100 without departing from the scope of the disclosed technology.

[0021] Players 108-110 may register for user accounts associated with gaming network 104. Such user accounts may allow users to access and play various media works on the gaming network platform. Gaming network 104 may also provide its registered users with options for verified accounts (e.g., professional, semi-professional, or other influencer users), team management, community tools, event tracking, social tools (e.g., friend lists), following other accounts, instant messaging, browsing live and archived streams, and other features related to playing games or watching games on the platform.

[0022] The game network 104 can be integrated with an event detection system 112 to allow automatic communication of game play data. The event detection system 112 can, for example, use an application programming interface (API) such as API 114 to receive the game play data. Subsequently, the game play data can be passed from API 114 to statistics module 116 for analysis. In a similar manner, game play statistics can be collected by API 114 and passed to statistics module 116. Depending on the implementation, game play statistics can be collected through active monitoring of the online game environment, or received from one or more external systems, for example, as part of a game performance data set that includes information describing individual player performance for one or more game titles and / or genre categories.

[0023] Because the metrics used to evaluate player performance against a particular game can vary widely between titles and genres, the ability to rank and seed players requires certain metrics to be considered. Such metrics can be converted from data generated from existing titles, or can be tracked on the platform during certain specified modes of gameplay.

[0024] In order to convert old game works, a conversion key provided by a game work developer or another party can be applied to a specific game performance attribute so as to evaluate the performance of different players associated with the game work in a standardized manner. For example, a conversion key can extract and evaluate certain metrics (e.g., number of scores, number of scores allowed, number of kills, match time) using appropriate weights applied to each metric. Old game works that rank each user using a basic score-based system can be mapped directly by the percentile of the ranked player. Certain predefined in-game events of interest can be associated with a set of game data indicating them. Such correlation can be provided by a developer, another party, or can be derived by applying artificial intelligence to a set of game data associated with a game work.

[0025] Such standardization also allows for objective comparison of skill levels, which can provide a basis for rankings. Different types of games, as well as different roles within each game, may involve different criteria. For example, team-based gameplay may involve teammates with different types of skill sets (e.g., different weapons, resources, in-game abilities, powers, etc.). In this way, gameplay of different team members may be evaluated according to different criteria, despite being played in the same game title.

[0026] Achievement or skill level titles can be provided based on the overall ranking of one or a group of players. For example, digital badges, trophies or medals can be provided to the highest-ranked players (or player teams) and published in the associated player (or team) profile. Such titles can be awarded in the current season and are still visible in later seasons. Different titles can be provided for different game works, different achievement combinations, overall ratings, etc. Such titles and grades can not only be used as the basis for matching opponents, but also the audience who wants to watch a player of a certain level in a certain game work can search for such indicators. Various search and classification options (for example, game works, rankings, stream parameters) can be provided so that players or potential audiences can filter available game streams, players, teams, tournaments, etc. to find other interested players. For example, players may wish to find a sparring partner with similar skills, or teammates with complementary skills. For example, such audiences can choose to pay attention to such players and request notifications, reminders and schedules related to the game involving interested players. In some embodiments, such titles can be made available for searching and filtering based on standardized skill evaluation systems through other platforms (for example, stream sources such as Twitch).

[0027] Different ranking systems may be applied to different game titles. In addition, different ranking systems may be used for tournaments. For example, a tournament may involve multiple game titles, and thus, the ranking system may combine and / or weight different individual rankings across game titles in a manner that may be customized for individual viewers or certain audiences. An overall ranking for any combination of games played on the platform may also be used. In this way, a player may distinguish himself as a skilled player not only in one or a few game titles (or even game series or genres), but in multiple different types of games.

[0028] Various audience tools that enable users to discover, tune in, watch tournaments, and interact with tournaments may take such rankings into account. Discovery may involve providing certain kinds of information (e.g., from player or team profiles) so that prospective viewers can determine whether a player, team, match, tournament, commentator, or other content producer / streamer may be of interest. Viewers (who may have their own user accounts) may also subscribe to or follow players, teams, game titles, or tournaments of interest. Thus, such subscribers may be provided with automatic notifications based on certain events or activities detected on the platform related to topics of interest (e.g., scores, lead changes, game results, standings after each round, updated pairings).

[0029] The event detection system 112 can also provide a way for viewers who have access to the game network platform 104 (e.g., via a user account) to discover digital content of interest based on detected events. Because registered viewers may already have user accounts, such discovery and detection can be based on user account information linked or otherwise accessed through the game network 104, which can also serve as the basis for broadcast / stream generation. For example, a user account can be associated with a group of friends (or fans, followers, or subscribers), which can serve as a basis for making suggestions, organizing audience viewing parties, or supporting interactions around the audience. For game works that can be played in teams, teams can be associated with team accounts or profiles including such data for use by the event detection system 112. Such team profiles can be linked to individual team member profiles.

[0030] Information about the progress or activity of each individual player or team within gaming network 104 may be automatically populated into the corresponding player or team profile. For example, a player profile may be tagged or include some other indicator that the player is playing a particular game, the rankings the player has achieved, a game schedule, opponents (and associated rankings), real-time game data related to the player, links to previously played game streams, links to active games, information about future scheduled games (e.g., games played in a tournament format), etc.

[0031] Furthermore, viewers who wish to observe a particular player may be linked to information related to a tournament stream involving the player of interest and other tournament information related to the player. Additionally, viewers who specifically wish to observe a tournament may be linked to an active tournament stream of the ongoing match and may be provided with the latest updates regarding the tournament.

[0032] In addition to viewers with clear interests (e.g., survey results, user account information, preferences, subscriptions within the gaming network platform 104), the event detection system 112 can further predict digital content that a particular viewer may be interested in. Additionally, events occurring within the digital environment can be identified as events of interest.

[0033] In addition, because the event detection system 112 is in communication with the gaming network 104, the event detection system 112 can receive and analyze gaming data in real time. In this way, the event detection system 112 can detect when a game starts, when a game is particularly active or competitive, and other notable in-game events. For example, certain in-game events can be associated with certain gaming data or metrics. The use of certain weapons, the display of certain character actions or maneuvers, a certain number or pattern of scores, etc. may indicate an abnormal or otherwise notable in-game event.

[0034] The event detection system 112 may apply elements of artificial intelligence and machine learning to game data obtained via the gaming network 104. As each game is played, additional game data may be added to the learning set to refine the ability of the event detection system 112 to identify digital content that may be of interest to a particular audience. Such game data may include not only in-game statistics (e.g., scores) explicitly captured by the game work, but may also include metrics that may be derived based on various analyses. In addition, the event detection system 112 may consider data associated with the game but external to the game. For example, such data may include sensor data (including accelerometers), biometric data, camera data, eye tracking data, comparisons to historical data or opponent data, Easter eggs, rarity or unusual nature of data (e.g., locations in rarely visited parts of the gaming environment, encounters with rarely encountered characters, acquisition of rare items or powers), current audience data (and their respective user profiles), etc.

[0035] It may be known that certain indicators are associated with events of interest, whether for a particular audience or in general. Such indicators may include the fame or degree of attraction of a player / team, the current number of audiences (e.g., audiences in a general audience or audience social circle) and growth, the amount of sharing or other activities related to the game, the ranking or record of a player / team, the excited gestures or actions (captured by a camera) of a player or current audience member, the excited chat in a chat room or other type of message board (e.g., hosted by a game network 104), the excited words (such as captured by a microphone), the biometric indicators of a player / team or audience (e.g., heart rate rises, pupil dilation), the instantaneity of winning, losing or breaking certain game records (e.g., the highest score in a game work on a game network 104), the abnormal actions or operations or use of power, the size of the ranking difference (e.g., the subversion of a low-ranking player / team to a high-ranking player / team). These factors may be evaluated relative to the timing of the game to identify the time period of the clusters that can focus on these factors. In addition, as the learning data set develops, certain data may be weighted and their weights may be adjusted over time.

[0036] When an event of interest is detected, a screenshot or video of such event can be captured and posted with a link that takes the viewer to the stream, either at the start of the game or directly to the point of interest. Such links can be accessed in real time (e.g., while the game is in progress) or in association with an archived stream. In some embodiments, such links can be posted on the gaming network 104 platform (e.g., a landing page, leaderboard, community, forum, user profile, team profile, tournament profile, bookmark) or at a location associated with a player, team, developer, broadcaster, other content producer or streamer or other viewer.

[0037] The identification of unusual, interesting, or otherwise significant events can also enable advertising, promotion, discovery, and related functionality. Notifications, alerts, video clips, screenshots, and other data can also be generated and provided in real-time to other parties, players, viewers, and potential viewers related to these events. Commentators and third-party broadcasters / streamers can also use such data to generate their own content streams.

[0038] The events identified in the game session can also be flagged, marked or otherwise tagged, and can be used for current and archived game streams. Moreover, such event identification can also be used to enhance the streams provided to the general audience or specific audience. Such enhancements can include pop-up notifications about various statistical information, comments or other discussion quotations, behind-the-scenes stories of players / teams / opponents, easy conversations, links, comparisons with history or other player / team records, immediacy of breaking records, comparisons with audience statistics, etc. The records considered can be historical records, game work records, personal records, tournament records, etc. on the game network 104. These flags can also be used to identify trends (e.g., player's progress), turning points or other long-term developments subsequently. Like this, flags can also be updated and improved over time.

[0039] Additionally, the view of a current or archived stream may be manipulated or re-rendered based on detected events. For example, certain in-game perspectives (e.g., selected players in a multiplayer game) may be highlighted or re-rendered from different angles, different viewing styles, time-slice effects, etc. using certain visual effects (including slow motion). In some embodiments, certain stream generators may modify streams for broadcast, which may include visual manipulations and edits. Data related to such modifications may be added to a learning set so that the event detection system 112 may learn which types of events may be of greater interest than others, perspectives or angles that may be more popular for certain events, types of visual effects or manipulations that are commonly applied to certain events, and the like.

[0040] Figure 22 is a block diagram of an exemplary stream selection engine that can be used in a system for discovering and detecting events that occur during the playback of interactive digital content. As shown, the stream selection engine 205 monitors multiple media streams associated with a game session 210 and selects streams that involve events of interest from popular angles or perspectives. Here, the media streams associated with the game session 210 include stream 1, stream 2, stream 3, and stream 4 (e.g., different player perspectives in a multiplayer game), and represent the display content or live game play in the game session 210. The display content can, for example, include individual player views (e.g., first-person perspective, third-person perspective, etc.), top views, bird's-eye views, two-dimensional (2D) views, 3 / 4 views, fixed three-dimensional (3D) views, omnidirectional views, God's-eye views, and other views that display the real-time game play of the game session 210.

[0041] The stream selection engine 205 also assigns one or more audience attributes 215 to each media stream, wherein each audience attribute 215 has a corresponding audience value. The audience attributes 215 may include any of the factors discussed above, including, for example, user votes, ratings (generally or among friends / followers), associated players or associated player attributes, historical popularity of view types or perspectives, and the like. As described herein, the terms "audience attributes" and / or "audience values" are synonymous and generally correspond to the interests, popularity, or preferences of each media stream. Audience values ​​may reflect current preferences or popularity, or may reflect trends within a given time period, such as changes or derivations of current preferences. Typically, higher or larger audience values ​​indicate stronger preferences; however, as will be appreciated by those skilled in the art, any value system may be used.

[0042] In operation, stream selection engine 205 monitors media streams 1-4 and continuously updates respective audience attributes / values ​​215 during real-time game play in game session 210 based in part on current user input. With regard to user input, multi-user platform and / or stream selection engine 205 provide user input interface including interactive graphic elements to request and receive input from current audience during real-time game play. Examples of audience input indicating interest or preference for a particular media stream include voting, ratings or other indications of preference for the media stream. As described below, one or more of these audience attributes may be considered independently or in combination with other audience attributes. In addition, as will be appreciated by those skilled in the art, some audience values / attributes represent weights applied to other audience values / attributes. Such audience input (from current audience) may be compared with user data associated with potential audiences to determine whether an event occurring during a game play session may be interesting.

[0043] Still reference Figure 2, stream selection engine 205 compares the single or gathered audience values ​​of each media stream during the real-time game, and determines the priority ranking of each media stream in the given time period of the real-time game. As discussed, stream selection engine 205 continuously updates respective audience values ​​215 during the real-time game based on current user input, and also compares or sorts the audience values ​​of corresponding media streams to determine the preference of a media stream viewed by the user relative to another media stream. Stream selection engine 205 also specifies one of the media streams to be broadcasted in a given time period based on its priority ranking at that time. The media stream specified in a given time period is also broadcast in audience channel 220 through the network. As shown in the figure, audience channel 320 represents the integration of the part (for a given time period) of each specified media stream. Stream selection engine 205 stitches or compiles a part of the currently specified media stream into audience channel 220. In the whole game session 210, stream selection engine 205 switches between media streams 1-4 based on the priority ranking of each media stream in a given time period. In this manner, viewer channel 220 provides user-preferred media content that displays different views of real-time game play during gaming session 210. Typically, a single media stream is designated to be broadcast per a given time period. However, it should also be appreciated that more than one media stream may be broadcast per a given time period.

[0044] Figure 3 is a flow chart illustrating an exemplary method 300 for discovering and detecting events in interactive digital content. Figure 3 The method 300 may be embodied as executable instructions in a non-transitory computer-readable storage medium, including but not limited to a CD, a DVD, or a non-volatile memory such as a hard drive. The instructions of the storage medium may be executed by one or more processors to cause various hardware components of a computing device that hosts or otherwise accesses the storage medium to implement the method. Figure 3 The steps shown in (and their order) are exemplary and may include various alternatives, equivalents or derivatives thereof, including but not limited to the order of execution thereof.

[0045] exist Figure 3 In the method 300, predefined event criteria may be maintained for each viewer, multiple games may be hosted on the gaming platform, game play data for each gaming session may be tracked, matches between the criteria and the tracked game play data may be identified, and streams may be filtered and classified based on such matches. Furthermore, the results of the filtering and classification may be provided to potential viewers.

[0046] In step 310, predefined criteria may be maintained in memory for certain in-game events. Some game events may have been defined in the game, such as scores. Other game events may become significant based on the intensity of such external information as the historical record of the game, the reputation or fame of the player / team, rarity, or current audience response. Some factors may be peculiar to the audience. For example, a record that is about to be broken may have belonged to a potential audience. Or the player who is about to break the record may be a friend of the audience. These factors may also be associated with certain weights that are peculiar to the audience.

[0047] Additionally, certain conditions may be identified as being common to known events of interest (e.g., archived streams from ended sessions). For example, trends regarding audience growth over the course of a session may be identified, as well as the volume and pattern of activity associated therewith. For example, shared patterns among current audience members may indicate that a particularly noteworthy event may be occurring.

[0048] In step 320, multiple games involving any number of players or teams of game play may be hosted on the gaming network 104 platform. The gaming session may involve different game titles available on the gaming network 104.

[0049] In step 330, data about each game session may be tracked in real time. Such data may include not only game play data within the game, but also analytics and metrics based thereon, spectator data, and external data related to the players / teams. Such games or matches are played on the gaming network 104 platform by players / teams with accounts or profiles, and viewed by spectators with profiles.

[0050] In step 340, certain events matching predefined event criteria may be identified within the gaming session. As described above, the event detection system 112 may evaluate the respective game play data and other data associated with the session in real time to identify noteworthy in-game events, which may include not only score or lead changes, but also unusual combinations of moves, significant speed, notable kills, underdog game overturns, broken records, etc.

[0051] In step 350, the available streams may be filtered and sorted according to the matches identified in step 340. In some cases, the available streams may include archived but unwatched streams as well as live streams. As described above, different weighting systems may be used to find the streams that are most likely to be of interest to the viewer.

[0052] The results of the filtering and sorting (e.g., a list of streams) may be displayed to the requesting party in step 360. The basis for such filtering and sorting (perhaps associated with the notability designations discussed herein) may also be displayed with the results, allowing the party (e.g., viewer) to jump not only to streams of interest, but directly to notable events depicted in the streams.

[0053] Figure 4 is an exemplary user device 400. User device 400 (eg, a desktop computer, laptop computer, tablet computer, mobile device, game console system) is a device that a user may use to facilitate execution of the third-party content viewing-related features of the present invention.

[0054] User equipment 400 may include Figure 4 It should be noted that the elements are exemplary and that other embodiments may incorporate more or fewer elements than those shown. Figure 4 , the user device 400 includes a main memory 402, a central processing unit (CPU) 404, at least one vector unit 406, a graphics processing unit 408, an input / output (I / O) processor 410, an I / O processor memory 412, a controller interface 414, a memory card 416, a universal serial bus (USB) interface 418, and an IEEE 1394 interface 420, an auxiliary (AUX) interface 422 for connecting a tracking device 424, but other bus standards and interfaces may be utilized. The user device 400 also includes an operating system read-only memory (OS ROM) 426 connected to the I / O processor 410 via a bus 434, a sound processing unit 428, an optical disk control unit 430, and a hard disk drive 432. The user device 400 also includes at least one tracking device 424.

[0055] The tracking device 424 can be a camera that includes eye tracking capabilities. The camera can be integrated into the user device or attached to the user device 400 as a peripheral device. In a typical eye tracking device, infrared non-collimated light is reflected from the eye and sensed by a camera or optical sensor. The information is then analyzed to extract eye rotation from the changes in reflection. Camera-based trackers focus on one or both eyes and record the movement of the eyes while the observer looks at a certain type of stimulus. Camera-based eye trackers use the center of the pupil and light to generate a corneal reflection (CR). The vector between the pupil center and the CR can be used to calculate the gaze point on the surface or gaze direction. Before using the eye tracker, a simple calibration step is usually required for the observer.

[0056] Instead, more sensitive trackers use reflections from the front of the cornea and the back of the eye's lens as features to track over time. Even more sensitive trackers image features from inside the eye, including retinal blood vessels, and follow these features as the eye turns.

[0057] Most eye tracking devices use a sampling rate of at least 30Hz, but 50 / 60Hz is most common. Some tracking devices run at up to 1250Hz, which is necessary to capture details of very fast eye movements.

[0058] Alternatively, a rangefinder camera can be used with the present invention to capture gestures made by the user and enable facial recognition. Rangefinder cameras are often used to capture and interpret specific gestures, which allow hands-free control of entertainment systems. This technology can use infrared projectors, cameras, depth sensors and microchips to track the movement of objects and individuals in three dimensions. This user device can also use a variation of image-based three-dimensional reconstruction.

[0059] Tracking device 424 may include a microphone, which is integrated into the user device or attached as a peripheral device to user device 400, which captures voice data. The microphone may perform sound source localization and / or ambient noise suppression.

[0060] Alternatively, the tracking device 424 can be a controller for the user device 400. When pointed at an LED in a sensor near, attached to, or integrated into the console of the entertainment system, the controller can sense its position in 3D space using a combination of built-in accelerometers and infrared detection. This design allows the user to control functions of the user device 400 through physical gestures as well as button presses. The controller connects to the user device 400 using wireless technology that allows data exchange over short distances (e.g., 30 feet). The controller may additionally include a "rumble" feature (i.e., shaking of the controller during certain points in the game) and / or an internal speaker.

[0061] The controller may additionally or alternatively be designed to capture biometric readings using sensors in the remote control to record data including, for example, skin moisture, heart rate, and muscle movement.

[0062] As described above, user device 400 may be an electronic game console. Alternatively, user device 400 may be implemented as a general purpose computer, a set-top box, or a handheld gaming device. In addition, similar user devices may include more or fewer operating components.

[0063] The CPU 404, the vector unit 406, the graphics processing unit 408, and the I / O processor 410 communicate via the system bus 436. In addition, the CPU 404 communicates with the main memory 402 via a dedicated bus 438, and the vector unit 406 and the graphics processing unit 408 can communicate via a dedicated bus 440. The CPU 404 executes programs stored in the OS ROM 426 and the main memory 402. The main memory 402 may contain pre-stored programs and programs transferred from a CD-ROM, DVD-ROM, or other optical disk (not shown) through the I / O processor 410 using the optical disk control unit 432. The I / O processor 410 mainly controls data exchange between various devices of the user device 400, including the CPU 404, the vector unit 406, the graphics processing unit 408, and the controller interface 414.

[0064] The graphics processing unit 408 executes graphics instructions received from the CPU 404 and the vector unit 406 to generate images for display on a display device (not shown). For example, the vector unit 406 may transform an object from three-dimensional coordinates to two-dimensional coordinates and send the two-dimensional coordinates to the graphics processing unit 408. In addition, the sound processing unit 430 executes instructions to generate sound signals, which are output to an audio device such as a speaker (not shown).

[0065] The user of user device 400 provides instructions to CPU 404 via controller interface 414. For example, the user may instruct CPU 404 to store certain information on memory card 416, or instruct user device 400 to perform a certain specified action.

[0066] Other devices may be connected to the user device 400 via the USB interface 418, the IEEE 1394 interface 420, and the AUX interface 422. Specifically, a tracking device 424 including a camera or a sensor may be connected to the user device 400 via the AUX interface 422, and a controller may be connected via the USB interface 418.

[0067] The present invention may be implemented in applications that may be run using a variety of devices. Non-transitory computer-readable storage media refers to any one or more media that participate in providing instructions to a central processing unit (CPU) for execution. Such media may take many forms, including but not limited to non-volatile and volatile media, such as optical or magnetic disks and dynamic memory, respectively. Common forms of non-transitory computer-readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROM disks, digital video disks (DVDs), any other optical media, RAM, PROM, EPROM, FLASHEPROM, and any other memory chip or cartridge.

[0068] Various forms of transmission media may be involved in carrying one or more sequences of one or more instructions to the CPU for execution. The bus carries the data to the system RAM, from which the CPU retrieves and executes the instructions. The instructions received by the system RAM may optionally be stored on a fixed disk before or after execution by the CPU. Similarly, various forms of storage may also be implemented, as well as the necessary network interfaces and network topologies to achieve the same purpose.

[0069] The detailed description of the technology has been provided above for illustration and description purposes. It is not intended to be exhaustive or to limit the technology to the precise form disclosed. Many modifications and variations are possible in light of the above teachings. The described embodiments are selected to best explain the principles of the technology, its practical application, and to enable others skilled in the art to use the technology in various embodiments and with various modifications to suit a particular intended use. It is intended that the scope of the technology be defined by the claims.

Claims

1. A system for discovering and detecting events in interactive digital content, the system comprising: at least one gaming network server that hosts a plurality of gaming titles and tracks real-time game play data in sessions involving the plurality of gaming titles, each session being associated with a stream; At least one server that: storing one or more event criteria for each of a plurality of predefined events of interest, wherein the event criteria include a time period for comparison with historical data to identify clusters of concentrated factors; identifying an event when at least one of the tracked gameplay data for the identified time period of one of the sessions meets the stored criteria; sorting a list of available flows identified as including the identified event, wherein a subset of the available flows is identified as being at the top of the list; providing a notification to a viewer device, the notification including a link to one of the streams in the subset; and updating the stored criteria based on use of a link to access a portion of the linked flow corresponding to the at least one identified time period; and an archive database storing the linked stream, wherein the portion of the linked stream corresponding to the at least one identified time period is marked as relating to the identified event, and the at least one server re-renders the portion of the linked stream and subsequent access to the linked stream includes the re-rendered portion.

2. The system of claim 1, wherein the event criteria includes data associated with one or more viewers.

3. The system of claim 1, wherein the event criteria includes a number of current viewers.

4. The system of claim 1, wherein the event criteria includes sensor data associated with at least one player or spectator.

5. The system of claim 1, wherein the event criteria includes an amount of social or shared activity.

6. The system of claim 1, wherein the at least one server further assigns weights to one or more event criteria and further sorts the list based on the assigned weights.

7. The system of claim 1, wherein the at least one server further enhances the linked stream based on the identified event.

8. The system of claim 7, wherein the linked stream is enhanced to include a pop-up notification.

9. The system of claim 1, wherein the at least one server identifies patterns of activity as common to predefined event types.

10. A method for discovering and detecting events in interactive digital content, the method comprising: Tracking real-time gameplay data across sessions involving multiple game titles, with each session associated with a stream; storing one or more event criteria for each of a plurality of predefined events of interest, wherein the event criteria include a time period for comparison with historical data to identify clusters of concentrated factors; identifying an event when at least one of the tracked gameplay data for the identified time period of one of the sessions meets the stored criteria; sorting a list of available flows identified as including the identified event, wherein a subset of the available flows is identified as being at the top of the list; providing a notification to a viewer device, the notification including a link to one of the streams in the subset; storing the linked stream in an archival database and marking a portion of the linked stream corresponding to the at least one identified time period as relating to the identified event; re-rendering the portion of the linked stream, wherein subsequent accesses to the linked stream include the re-rendered portion; as well as The stored criteria are updated based on use of the link to access the portion of the linked flow corresponding to the at least one identified time period. The method of claim 10 , wherein the event criteria includes data associated with one or more viewers. The method of claim 10 , wherein the event criteria includes a number of current viewers.

13. The method of claim 10, wherein the event criteria includes sensor data associated with at least one player or spectator. The method of claim 10 , wherein the event criteria includes an amount of social or shared activity.

15. The method of claim 10, further comprising assigning weights to one or more event criteria, wherein the list is further sorted based on the assigned weights.

16. The method of claim 10, further comprising enhancing the linked stream based on the identified event.

17. The method of claim 16, wherein the linked stream is enhanced to include a pop-up notification.

18. The method of claim 10, further comprising identifying patterns of activity as common to predefined event types.

19. A non-transitory computer-readable storage medium having embodied thereon a program executable by a processor to perform a method for discovering and detecting events in interactive digital content, the method comprising: Tracking real-time gameplay data across sessions involving multiple game titles, with each session associated with a stream; storing one or more event criteria for each of a plurality of predefined events of interest, wherein the event criteria include a time period for comparison with historical data to identify clusters of concentrated factors; identifying an event when at least one of the tracked gameplay data for the identified time period of one of the sessions meets the stored criteria; sorting a list of available flows identified as including the identified event, wherein a subset of the available flows is identified as being at the top of the list; providing a notification to a viewer device, the notification including a link to one of the streams in the subset; storing the linked stream in an archive database and marking a portion of the linked stream corresponding to the at least one identified time period as relating to the identified event; re-rendering the portion of the linked stream, wherein subsequent accesses to the linked stream include the re-rendered portion; as well as The stored criteria are updated based on use of the link to access the portion of the linked flow corresponding to the at least one identified time period.

Images