• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    System and Method for Optical Tracking of Players in Sports Venues A system and method is provided for optically tracking players and automatically capturing irradiable video of a sporting event within a venue. The system includes at least one tracking fixed camera, a remote-controlled broadcast camera, and a scoreboard, with each component communicating with a processing system. The method involves dividing the site into zones of interest, optically tracking individual players in the zones of interest, identifying groups of players in the zones of interest, which determines a zone of interest that contains the majority of players by analyzing player activity in the zones of interest. interest to most players, selecting a current gameplay scenario corresponding to the player activity of a set of predetermined game scenarios, and adjusting the pan, tilt and zoom of a broadcast camera to match the current game scenario .
    公开号:BR112017007408A2
    申请号:R112017007408-7
    申请日:2015-05-08
    公开日:2019-11-19
    发明作者:Miller Farrel；ibrahim Peter
    申请人:Liverbarn Inc；
    IPC主号:
    专利说明:

    SYSTEM AND METHOD FOR OPTICAL TRACKING OF PLAYERS IN
    SPORTS STADIUMS
    FIELD OF THE INVENTION [001] The present invention relates to optical tracking, and more specifically to optical tracking based on zones to automatically capture a transmissible video of a sporting event inside a stadium.
    FUNDAMENTALS [002] Recording and broadcasting sporting events is a very common practice at the professional level. Spectators are able to experience events that they could not otherwise attend in person, and to review exciting moments from past events. This opens the event up to a much larger audience, increasing the potential audience and benefiting both the team and the league. Spectators watching the event live can also have their experiences enhanced by gaining additional observation points and being able to review and study important games in order to improve their tactics.
    [003] Typically, a quality broadcast of a sporting event requires a significant amount of equipment and manpower. Depending on the type of event, 20 or more cameras may be scattered around the stadium, many of which must be operated manually by a human videographer to follow the action of the game. The images from each camera are received in a central control room where several operators select the appropriate images to better portray the action taking place in the game. Images must be edited on the fly to produce a transmissive product! that viewers can follow in their
    Petition 870170035862, of 05/29/2017, p. 6/50
    2/34 televisions, computers or mobile devices.
    [004] Disadvantageously, the high cost of producing a quality transmission makes such practice inaccessible at lower levels of sports. Less professional leagues resort to the use of few cameras and with a lower transmission quality. Semi-professional, junior, youth and amateur leagues often cannot afford to record or broadcast games, in any way leaving the responsibility for individual spectators and coaches to film games if they wish to review and study them.
    [005] Various efforts have been made to make the broadcast of sporting events more accessible. Most of these efforts involve automating most of the process in order to reduce the amount of human labor required. In particular, specialized systems have been developed that use computer-controlled cameras to track play in a game and automatically generate transmissible images.
    [006] For example, US 2010/0026809 discloses the use of transceivers, accelerometers, transponders and / or detectable RADAR elements embedded in players 'equipment to individually track players' locations in three dimensions. Positioning information is transmitted to a computer system that controls the horizontal movement, vertical movement and zoom of 2D cameras in various positions to capture the gameplay.
    [007] Another example is US 6,567,116 which also reveals the tracking of players in three dimensions.
    [008] The system uses the use of plasters, adhesives or
    Petition 870170035862, of 05/29/2017, p. 7/50
    3/34 frequency selective reflective tapes attached to the player's equipment in order to act as a marker. Tracking cameras scattered throughout the stadium detect the reflected radiation in order to identify and locate marked objects and players. The position information is subsequently supplied to a computer system that can separate the marked objects from their background. A related patent, US 7,483,049, discloses the use of such a tracking system to target multiple fixed cameras with adjustable side view to capture and record the game event.
    [009] State of the art systems are disadvantageously still too expensive to be accessible to all levels of sports. Although less labor is required to operate these systems, they require additional equipment, such as multiple tracking cameras, built-in tracking devices, or reflective patches, which increases the cost and complexity of the system. Additionally, state-of-the-art fully automated systems rely on the viewer to choose a desired viewing angle and therefore cannot produce a high quality video broadcast that gives the best view of a game without human intervention.
    [010] There is therefore a need for an improved player tracking system for use in sports stadiums that reduces the total cost while being able to automatically produce a high-quality video broadcast of a sporting event without human intervention and at a cost. reduced equipment.
    Petition 870170035862, of 05/29/2017, p. 8/50
    4/34
    SUMMARY [Oil] In accordance with one aspect of the invention, a method is provided to automatically capture transmissible video from a sporting event inside a stadium. The method includes the steps of: dividing the stadium into a plurality of zones of interest; optically track individual players in the stadium by determining the respective positions of individual players within the zones of interest using at least one fixed tracking camera; based on the respective positions of the individual players, perform a cluster analysis to identify groups of players within the zones of interest; determine a particular zone of the zones of interest that contains the largest number of individual players; analyze the activity of individual players and groups of players within the given zone of the zones of interest to select a current gameplay scenario corresponding to that activity, the current gameplay scenario being selected from a predetermined set of gameplay scenarios; based on the current gameplay scenario, adjust the horizontal movement, vertical movement and zoom control of a broadcast camera in real time, thereby changing a field of view of the broadcast camera and capturing the broadcast video of the sporting event.
    [012] In a modality, the current gameplay scenario is selected based on the current position, speed or direction of at least one of the individual players or at least one of the groups of players in relation to the determined zone of the zones of interest.
    [013] In one embodiment, the method includes identifying
    Petition 870170035862, of 05/29/2017, p. 9/50
    5/34 individual players or groups of players that transit between adjacent zones of interest, and select the current gameplay scenario based on the current position, speed or direction of the individual players or groups of players that transit between the adjacent zones.
    [014] In one modality, the method includes predicting a transition for individual players or groups of players between adjacent first and second zones, and analyzing the activity of individual players or groups of players within both the first and second zones with the purpose of selecting the current gameplay scenario.
    [015] In one mode, optically tracking individual players includes the steps of extracting foreground mosaics from tracking video frames captured by at least one fixed tracking camera; determine locations of foreground mosaics based on their positions within the frames of the tracking video; correlate each of the foreground mosaics to a respective player of the individual players; and maintaining a list of tracked players, each player tracked on the list being associated with a predetermined number of the player's last known locations.
    [016] In one embodiment, correlating foreground mosaics includes matching the location of a foreground mosaic to a player on the list that has a known last location closest to the foreground mosaic, and if a match is found, updating the list so that the last known location
    Petition 870170035862, of 05/29/2017, p. 10/50
    6/34 associated with the player matches the location of the foreground mosaic, and if no match is found, add a new tracked player to the list with a last known location corresponding to the location of the foreground mosaic.
    [017] In one embodiment, extracting the foreground mosaics from the tracking video frames captured by at least one fixed tracking camera includes subtracting sequential frames from the tracking video to obtain a difference image, and extracting the first mosaics difference image plane by performing the steps of: applying a limit to the difference image to obtain a binary image of the foreground mosaics on a background; obscuring the binary image to remove noise; and run a contour algorithm on the binary image to extract the mosaics.
    [018] In one modality, the method includes identifying mosaics in movement corresponding to players in movement and mosaics stationary corresponding The players stationary. Moving mosaics are extracted of image of difference and the mosaics stationary are extracted in a single frame from the video tracking.[019] In a mode, select the current scenario in
    Gameplay includes matching the activity of individual players and groups of players to a predetermined pattern associated with a specific gameplay scenario in the predetermined set of gameplay scenarios.
    [020] In one mode, the method involves dividing the stadium so that the zones of interest correspond to zones on a playing field within the stadium.
    Petition 870170035862, of 05/29/2017, p. 11/50
    7/34 [021] In one embodiment, the method includes the step of calibrating at least one fixed tracking camera by mapping a tracking region on a playing field within the stadium, the tracking region being a region within the which the tracking video is used to optically track individual players, and outside which the tracking video is ignored.
    [022] In one embodiment, individual players are tracked using a plurality of fixed tracking cameras, and the method includes the steps of normalizing tracking data through the plurality of tracking cameras, and merging the tracking data of the entire plurality tracking cameras for the purpose of
    track back the individual players globally inside Stadium. [023] In a modality, the method includes performing a analyze in grouping on the position of players
    to identify a global group of players within the stadium, and select the gameplay scenario based on the current position, speed or direction of the global group.
    [024] In one mode, the method includes the steps of capturing the scoreboard video and combining the referred scoreboard video over the transmissible video, the scoreboard video comprising a scoreboard video inside the stadium.
    [025] In one mode, the method includes the step of adjusting the horizontal movement, vertical movement and zoom control of the broadcasting camera so that the field of view is limited to containing one player to the left, one player to the right , one player higher and one player
    Petition 870170035862, of 05/29/2017, p. 12/50
    8/34 further down within the determined zone of the zones of interest or within the stadium.
    [026] In one embodiment, the method includes the step of transmitting the transmissible video as live images over a network.
    [027] An additional modality of the method may include any combination of the aforementioned modalities.
    [028] In accordance with one aspect of the invention, a system is provided to automatically capture transmissible video from a sporting event inside a stadium. The system includes at least one fixed tracking camera, a transmission camera and a processing system. At least one fixed tracking camera is configured to capture tracking video of players in the stadium. The broadcast camera is configured to capture the transmissible video of the sporting event, and is remotely controlled to adjust its horizontal movement, vertical movement and zoom control. The processing system is operatively coupled to at least one tracking camera and the transmission camera. The processing system is configured to: divide the stadium into a plurality of zones of interest; optically track individual players in the stadium by determining the respective positions of individual players within the zones of interest using at least one fixed tracking camera; based on the respective positions of the individual players, perform a cluster analysis to identify groups of players within the zones of interest; determine a particular zone of the zones of interest that contains the largest number of players
    Petition 870170035862, of 05/29/2017, p. 13/50
    9/34 individual; analyze the activity of individual players and groups of players within the given zone of the zones of interest to select a current gameplay scenario corresponding to that activity, the current gameplay scenario being selected from a predetermined set of gameplay scenarios; based on the current gameplay scenario, adjust the horizontal movement control, vertical movement and zoom of the broadcast camera in real time, thereby changing a field of view of the broadcast camera and capturing the transmissible video of the sporting event.
    [029] In one embodiment, there are a plurality of tracking cameras, and each of the plurality of tracking cameras is designed to capture tracking video of a particular zone from the plurality of zones of interest in the stadium.
    [030] In one embodiment, at least one tracking camera includes an optical tracking module. The optical tracking module is configured to: extract foreground mosaics from frames of tracking video captured by at least one fixed tracking camera; determine locations of foreground mosaics based on their positions within the frames of the tracking video; and transmitting coordinates of the locations of the foreground mosaics to the processing system, said coordinates being transmitted without also transmitting the tracking video to the processing system.
    [031] In one embodiment, the system includes a scoreboard camera operatively coupled to the processing system,
    Petition 870170035862, of 05/29/2017, p. 14/50
    10/34 the scoreboard camera being set up to capture scoreboard video from a scoreboard inside the stadium. The processing system is configured to combine the scoreboard video with the transmissible video.
    [032] An additional modality of the system may include any combination of the above mentioned modalities.
    [033] According to one aspect of the invention, a machine-readable non-transient medium is provided that executes one or more sequences of instructions. When executed by one or more processors operatively coupled to at least one tracking camera and a remotely controllable broadcast camera in a stadium during a sporting event, the instructions cause the one or more processors to perform the steps of: dividing the stadium in a plurality of zones of interest; optically track individual players in the stadium by determining the respective positions of individual players within the zones of interest using at least one fixed tracking camera; based on the respective positions of the individual players, perform a cluster analysis to identify groups of players within the zones of interest; determine a particular zone of the zones of interest that contains the largest number of individual players; analyze the activity of individual players and groups of players within the given zone of the zones of interest to select a current gameplay scenario corresponding to that activity, the current gameplay scenario being selected from a predetermined set of gameplay scenarios; based on the current gameplay scenario, adjust the
    Petition 870170035862, of 05/29/2017, p. 15/50
    11/34 horizontal movement, vertical movement and zoom of the transmission camera in real time, thus changing a field of view of the transmission camera and capturing the transmissible video of the sporting event.
    BRIEF DESCRIPTION OF THE DRAWINGS [034] Figure 1 is a schematic diagram of a system for automatically capturing transmissible video of a sporting event inside a stadium, according to one modality.
    [035] Figure 2 is a schematic diagram of a tracking camera used in the system in Figure 1.
    [036] Figures 3A and 3B are schematic diagrams that illustrate the mapping of areas of interest on a playing field.
    [037] Figure 4 is a flowchart that illustrates a method for automatically capturing transmissible video of a sporting event inside a stadium, according to one modality.
    [038] Figure 5 is a flow chart illustrating a method for optically tracking individual players, useful in the method in Figure 4.
    [039] Figure 6 is a flowchart that illustrates a method for extracting foreground mosaics from tracking video frames, useful in the method of Figure 5.
    [040] Figure 7 is a flowchart that illustrates a method for identifying a gameplay scenario corresponding to a breakaway, useful in the method in Figure 4.
    [041] Figure 8 is a flowchart that illustrates a method for selecting a gameplay scenario from predetermined gameplay scenarios and for adjusting the
    Petition 870170035862, of 05/29/2017, p. 16/50
    12/34 horizontal movement, vertical movement and zoom values for the transmission camera, useful in the method of Figure 4.
    [042] Figures 9A, 9B and 9C are schematic diagrams that illustrate the adjustment of the horizontal movement, vertical movement and zoom values of the transmission camera to change its field of view.
    [043] Figure 10 is a schematic diagram that illustrates the scoreboard video composed over the transmissible video of the sporting event.
    DETAILED DESCRIPTION [044] The following describes preferred embodiments of aspects of the present invention. These are some of the many ways to implement the invention. As such, the examples provided should in no way be considered to limit the scope of the invention.
    System [045] In accordance with one aspect of the invention, a system is provided to automatically capture transmissible video from a sporting event inside a stadium. The system is configured to track players inside a stadium, record game action, and transmit video footage of the game's action to an audience. With reference to Figure 1, a modality of system 100 is illustrated schematically. System 100 is installed in a sports stadium 101, in this case a hockey arena. Other types of sports stadiums are also possible, such as football, soccer or baseball stadiums, for example. Sports stadium 101 includes a playing field 102, in this case a hockey rink. In other modalities, other types of playing fields are possible, such as a basketball court,
    Petition 870170035862, of 05/29/2017, p. 17/50
    13/34 or football, american football or baseball fields.
    Typically, sports stadiums 101 include other elements such as spectators and rest areas for players. They can also include a 109 scoreboard to visually display the game score.
    [046] System 100 includes at least one fixed tracking camera 200 facing the playing field 102. It is said that at least one tracking camera 200 is fixed due to the fact that it is positioned and calibrated in a specific location , and is not designed to move. In other words, the field of view (FOV) of at least one tracking camera is preferably static, due to the fact that at least one tracking camera cannot have horizontal or vertical movement. In the illustrated mode, there are three fixed tracking cameras 200, but in other modes more or less tracking cameras can be used. In the present modality, the tracking cameras 200 are equally spaced on one side of the playing field 102. They are also positioned above and away from the edge of the playing surface at a distance that obtains the optimum coverage based on the horizontal and vertical FOVs of the video cameras. tracking 200. In the present mode, each tracking camera preferably covers approximately one third of the playing field 102, however there may be overlaps in its fields of view.
    [047] With reference to Figures 3A and 3B, tracking cameras 200 can be positioned and calibrated to cover a specific tracking region 300 in the stadium. In the present modality, the tracking cameras 200 are positioned and calibrated in order to map a region
    Petition 870170035862, of 05/29/2017, p. 18/50
    14/34 tracking 300 on the playing field 102. When doing this, objects are only tracked if they are located within the tracking region 300. Objects outside the tracking region 300, for example spectators in the stands or players in bank 304, can be ignored or filtered out.
    [048] Tracking cameras 200 can also be configured so that they are individually designed to capture tracking video from a specific area of interest. In the present embodiment, the playing field 102 is divided into three zones of interest 301, 302 and 303. Each of these zones of interest corresponds to a zone of the playing field. In the present modality, the first zone of interest 301 corresponds to the defensive zone of the hockey ring; the second zone of interest 302 corresponds to the neutral zone; and the third zone of interest 303 corresponds to the offensive zone. Of course, other zone divisions are also possible. For example, other areas of interest could correspond to game restart circles. Each of the 200 ', 200 and 200' tracking cameras is positioned to capture tracking video mainly in its designated area of interest. If the camera's FOVs overlap, each camera can be individually calibrated to ignore movement within its field of view if that movement occurs outside the camera's designated zone of interest.
    [049] The scope of the invention allows that in other modalities the number of cameras and their positions can vary. For example, in a given playing field, as little as 1 or as many as n cameras can be used
    Petition 870170035862, of 05/29/2017, p. 19/50
    15/34 tracking. For n cameras, each camera can cover approximately 1 / n of the field, with possible overlap between each FOV of the camera. Additionally, the spacing between each tracking camera does not necessarily need to be equidistant. Some cameras may be closer together, while others are more spaced from each other. For example, multiple cameras can be grouped together for the purpose of tracking heavily tracked areas of the field, while fewer cameras may be more dispersed among themselves to track less tracked areas. Of course, when the spacing between cameras is not equidistant, the FOV overlap between each camera can vary widely.
    [050] The function of tracking cameras 200 is to provide the system with information regarding the movements of players on the playing field 102. With reference to Figure 2, each tracking camera unit 200 consists of a video camera 202 that includes a module tracking video capture 204, an optical tracking module 206, and a communications module 208. The video camera can be configured, for example, to capture video in the visible spectrum, the infrared spectrum, or both. The tracking video capture module 204 is responsible for capturing tracking video using a sensor provided in the tracking camera 200. The optical tracking module 206 receives a video signal from the capture module 204, and executes video vision algorithms. machine with the purpose of tracking mosaics in the video signal. In the present mode, the output of the optical tracking module 206 consists of a 3-tuple of x and y position and dimension of identified mosaics
    Petition 870170035862, of 05/29/2017, p. 20/50
    16/34 in the video. This double x and y can be called the location or coordinates of the mosaics. In other embodiments, however, coordinates can include other parameters, such as a z position or orientation, for example. The output of the tracking module 206 is transmitted via the communication module 208 over a wired or wireless connection 210 to a processing system for further processing. It should be noted that in the present modality, only the 3 coordinate set is transmitted by connection 210, and not the tracking video itself, reducing the amount of bandwidth required.
    [051] In the present embodiment, the single tracking camera unit 200 includes the capture module 204, the tracking module 206 and the communication module 208. In other words, the processing of the raw tracking video images and the Optical mosaic tracking takes place locally on camera 200. These modules can be an integral part of a video camera 202, or they can be part of a single-board computer attached to a video camera 202, for example.
    [052] Referring again to Figure 1, system 100 also includes two additional cameras to capture video of the sporting event. The first is a broadcast camera 106 which is a video camera positioned near the center of the field, preferably on the same side as the tracking cameras. The transmission camera 106 is mobile and can be controlled remotely via an application programming interface (API) to adjust its horizontal movement, vertical movement and zoom (PTZ). According to possible modalities, PTZ can be adjusted by moving
    Petition 870170035862, of 05/29/2017, p. 21/50
    17/34 physically the camera or lens through actuators, or digitally by applying a video effect. The broadcast camera 106 serves to capture video images of the action at the sporting event. The second additional camera is a scoreboard camera 108. The scoreboard camera 108 is a video camera aimed at score 109 at stadium 101. It serves to capture video images from scoreboard 109 which can later be composed on top of the video images. from the broadcast camera.
    [053] Tracking cameras 200, transmission camera 106 and scoreboard 108 are all in communication with a processing system 110. In the present embodiment, the processing system is a computing unit 110, such as a server , for example. Communication with the computing unit 110 can be carried out over a local network, and can be either a wired or wireless connection. The computing unit 110 includes at least one processor and memory. The computing unit is configured to receive tracking data from each of the tracking cameras 200 in the form of x, y and dimension data. The computing unit 110 then executes a method implemented in software on the tracking data to determine the gameplay, and then controls the PTZ of the transmission camera 106 to obtain the best view of the game action. The computing unit 110 can also receive video images from both the broadcast camera 106 and the scoreboard 108, and can create a composition from those to show both the game action and the current score. The resulting composite image can be further manipulated to add other effects or
    Petition 870170035862, of 05/29/2017, p. 22/50
    18/34 graphics before being distributed or transmitted over a network 111, such as the internet, for example, to client devices 112 such as computers, smart phones, tablets or other similar devices.
    [054] In some modalities, the video can be transmitted over the internet through a content distribution network (CDN). In such a mode, the video captured by the transmission camera 106 and the scoreboard camera 108 is transmitted to the CDN. The video can be transmitted either directly from the cameras, or it can be transmitted via computer unit 110. CDN is then responsible for transcoding the video for compatibility with various platforms, and distributing the video to the client's 112 devices. The video it is delivered to native applications that run on client devices 112, and can also be made available through a web application. In some modalities, the video can be distributed to client devices 112 in the form of live video images of the sporting event.
    [055] System 100 can also include other devices or sensors to further improve the transmitted video. For example, the system can be equipped with a microphone or a set of microphones to capture audio from the sporting event and the audience. The microphones can be directional or non-directional and can be placed strategically throughout the stadium. In some embodiments, the transmission camera 106 may be provided with a directional microphone such that audio in the field of view of the transmission camera 106 is captured. The system can also include a microphone capable of controlling horizontal movement,
    Petition 870170035862, of 05/29/2017, p. 23/50
    19/34 vertical movement and zoom that can be controlled by the processing system 110 in a similar way to the transmission camera 106 in order to follow the action of the game.
    [056] An advantage of the present invention is that it allows dimensioning other modalities according to different needs. For example, in other modalities, the system may comprise multiple transmission cameras positioned at other advantageous points around or above the playing field. In addition, more tracking cameras can be used for larger fields, and less can be used for smaller fields. In some embodiments, the computing unit may be located on site. In other embodiments, the computing unit may be off-site and may be cloud-based to further reduce the amount of hardware involved. According to possible modalities, each of the cameras can be HD, or of higher resolutions such as 2K or 4K. Finally, tracking cameras and broadcast cameras can be integrated into a single unit to further reduce costs and the amount of equipment involved.
    Method [057] In accordance with one aspect of the invention, a method is provided to automatically capture transmissible video from a sporting event within a stadium. The method includes steps to track players, record game action, and transmit video footage to an audience. The method is preferably carried out on a system such as that described above.
    [058] Briefly described, and with reference to Figure 4, the method includes the steps of: dividing the stadium into zones
    Petition 870170035862, of 05/29/2017, p. 24/50
    20/34 of interest 402; optically track individual 500 players; identify groups of players in areas of interest 40 6; determine an interest zone that contains the most 408 players; analyze the activity in the area of interest with the greatest number of 410 players; select a gameplay scenario corresponding to activity 412; and adjust a horizontal movement, vertical movement and zoom control according to the 414 gameplay scenario. The steps between 500 and 414 are repeated continuously, so that players are continuously tracked and the broadcast camera's PTZ is updated in time real.
    [059] With reference to Figure 5, the step of optically tracking individual players 500 may include the steps of: capturing tracking video 502; extract foreground mosaics from the frames of the tracking video 600; determine 504 foreground tile locations; transmitting mosaic locations to a 506 processing system; merge tracking data from multiple 508 tracking cameras; correlate mosaics to individual 510 players; and maintain a list of 512 tracked players. The steps can be repeated continuously.
    [060] Many of these steps can be performed in a central processing system, while some steps can be downloaded to other dedicated modules, such as a dedicated module in the tracking cameras. In one embodiment, the method may first involve analyzing the video data in a tracking camera unit to determine x, y, and mosaic size information for players or groups of players. Then, a processing unit receives information from
    Petition 870170035862, of 05/29/2017, p. 25/50
    21/34
    mosaics of the multiple camera units and runs one process implemented in software with goal in analyze the action of game. Finally, the system in processing controls a broadcast camera for to provide the best view of the game action. [061] In the modality currently described , a first
    portion of the method can be performed on each tracking camera unit. With reference to Figures 5 and 6, the optical tracking module of the tracking camera unit can perform the following steps:
    a) Read a video frame image; wait for a period of time before reading another video frame image, 502;
    b) Subtract the two sequential images to obtain an absolute difference image, 604;
    c) Apply a limit to convert the difference image to a binary image, 606;
    d) Apply an obscuration filter to the binary image to remove noise, 608;
    e) Apply another limit to revive the binary image, 610;
    f) Perform a contour process to extract the foreground mosaics from the background, 612;
    g) Discriminate mosaics based on their dimensions to remove unwanted noise and mosaics that are smaller than n pixels in dimension;
    h) Convert the absolute pixel x, y position of the mosaic based on the image width and height in normalized values 0.0-1.0, 504;
    i) Transmit the 3-tuple of position x, y and dimension of
    Petition 870170035862, of 05/29/2017, p. 26/50
    22/34 each mosaic to the processing system for use in the process implemented in software to determine gameplay,
    6; and
    j) Repeat from step a).
    [062] In one embodiment, the method may involve using a different image processing technique to extract static mosaics than that to extract moving mosaics. This can allow the system to better handle players who are not on the move. Referring to Figure 6, the method can include a step 602 to determine whether a mosaic is in motion. If the mosaic is in motion, steps a) to j) are performed normally as described above. If the mosaic is not moving, its location is determined by analyzing a single frame of the tracking video. In this way, the analysis is not performed on a difference image, and the step of subtracting sequential frames is ignored.
    [063] In the present embodiment, a second portion of the method can be performed in the processing system. The processing system's function is to receive player position information from tracking cameras, perform a process to determine gameplay, and control the broadcast camera's PTZ. Gameplay can be determined by analyzing player activity within a zone of interest and selecting a gameplay scenario from a predetermined set of gameplay scenarios. The set of scenarios may include, for example, a game restart, a breakaway, or player movement. A scenario can be selected based on a pattern of players or a sequence of events
    Petition 870170035862, of 05/29/2017, p. 27/50
    23/34 of players or groups of players in the zones of interest.
    Once a scenario is selected, the broadcast camera's PTZ is adjusted to provide the best view of the game's action according to the scenario.
    [064] In one mode, the following steps can be performed by a continuous loop processing system. In the modality currently described, the system includes four tracking cameras to track players in four zones of interest: the defensive zone (zone 1), half of the neutral zone closest to the defensive zone (zone 2), half of the neutral zone closest to the offensive zone (zone 3) and the offensive zone (zone 4).
    [065] The processing system receives data asynchronously from each of the tracking camera units. The data are in the form of a list of 3-tuple position x, y and dimension of each of the mosaics detected in the units of tracking cameras.
    [066] Each x, y position of the mosaic is normalized based on the width and height of the cameras' FOV pixels:
    - Mosaic position x = 0 means leftmost pixel of horizontal FOV of tracking camera;
    - Mosaic position x = 1.0, means rightmost pixel of horizontal FOV of tracking camera;
    - Mosaic position y = 0 means pixel more high in FOV vertical camera tracking; and - Y mosaic position = 1.0 means pixel more low in FOV vertical camera tracking.
    [067] In the present mode, the y position is the same for all four tracking cameras and is not modified. The x position needs to be normalized across all 4
    Petition 870170035862, of 05/29/2017, p. 28/50
    24/34 tracking cameras where:
    Camera 1, The position X varies in 0-0.25;Camera 2, The position X varies in 0.25-0, 50; Camera 3, The position X varies in 0.50-0, 75; and Camera 4, The position X varies in 0.75-1, 0.
    [068] The normalized mosaic data is placed in a list of working mosaics.
    [069] The processing system then performs the player tracking process that matches past data with current data (list of work tiles, or list of tracked players) to determine which tiles are the same, but shifted in space and time . The process filters the tiles based on dimension> n pixels and places them in a list. The value of n can be chosen based on the relative size of player tiles and calibrated based on the position and resolution of the tracking cameras. For example, the value of n can vary between 10 and 25 pixels. In an exemplary embodiment, it cannot be 10.7 pixels, considering that all tiles smaller than 10.7 square pixels in size are noisy. The mosaic tracking process uses a minimum distance calculation to find correspondence between past mosaics and mosaics in the list. If the distance calculation is less than a minimum limit then there is a match and the tile is assigned to a player object and added to the player tile list. The mosaic added to the list thus corresponds to the last known most recent location of a tracked player. Mosaics that do not match previous data are then considered to be new and assigned to a player object.
    Petition 870170035862, of 05/29/2017, p. 29/50
    25/34
    The list of player tiles is maintained to a maximum extent of the n tiles. In other words, in the list of tracked players, each tracked player is associated with a predetermined number n of their last known locations. The value of n in this context can be chosen based on the least amount of prior information needed to determine a player's direction of movement. The value of n can vary, for example, between 5 and 10 last tiles. In an exemplary mode, n is chosen as 5, so that each player tile list contains only 5 tiles corresponding to the player's last known locations.
    [070] The player tile list is used to calculate the player's speed and direction. The calculation of the player's horizontal or x direction is based on the loop through the player's tile list and the sum of the difference between the new x position and the previous x position.
    xDir = Σ7 = ο '* = (xDir [iH] -xDir [i]) [071] Similarly, the calculation of the player's vertical or y direction is based on the loop through the player's tile list and the sum of the difference between the new y position and the previous y position.
    yDir = Xjto 1 1 - (yDir [i4l] -yDir [i]) [072] A positive value of xDir implies that the player is moving within the tracking camera image from left to right. A negative xDir value implies that the player is moving within the image from the tracking camera from right to left. A zero value
    Petition 870170035862, of 05/29/2017, p. 30/50
    26/34 implies that the player is not moving in the horizontal direction. The magnitude of xDir is the speed in the horizontal direction.
    [073] A positive yDir value implies that the player is moving within the image from the tracking camera from top to bottom. A negative yDir value implies that the player is moving within the image from the tracking camera from the bottom up. A value of zero implies that the player is not moving in the vertical direction. The magnitude of yDir is the velocity in the vertical direction.
    [074] Simplified grouping can be applied to reduce processing complexity. In one embodiment, each tracking camera can be considered a potential grouping or group of players. A tracking camera can consist of a group of 0 or 1 group of players. Therefore, there can be a minimum of 0 groups of players and a maximum of 4 groups of players through all tracking cameras. In other modalities, players can be divided into different groups of players. Performing a cluster analysis in a zone of interest or across the playing field can allow the identification of groups of players within the zones of interest, or a global group of players within the entire stadium. Groupings can be further divided into subgroups.
    [075] If a tracking camera has detected players, the speed and direction of the group is calculated from the individual speed and direction of the players. Average the sum of all players moving to the left, averaging the sum of all players
    Petition 870170035862, of 05/29/2017, p. 31/50
    27/34 players moving to the right and considering the maximum value of the two averages, the speed and direction of the group are obtained. Speed and direction are calculated for all tracking cameras.
    [076] The leftmost player, the rightmost player, the highest player and the lowest player are determined based on the x and y position values of all players. The x positions of the players on the left and right and the y positions of the players above and below are used to calculate the potential FOV of the camera. A narrow FOV implies that players on the left and right have relatively close x positions or that players above and below have relatively close y positions. A wide FOV implies that the x and y positions are apart.
    [077] Potential FOV can be used in some gameplay scenarios to give an overview of all players on the playing field, or to give an overview of all players within a specific area of interest. As players move, the potential FOV changes so that it is limited to contain all players. In this way, the FOV calculated in this way provides a good view for regular gameplay when players move from one end of the ice to the other.
    [078] Potential FOV can be calculated so that it is limited to contain all players within a zone, or to contain all players in motion on the playing field.
    game. With reference at Figures 9A and 9B, a first FOV 902 can to be calculated in mode what be limited to to contain all the players inside gives zone offensive 303. In this configuration, the player 904 it's the more to the left; O player
    Petition 870170035862, of 05/29/2017, p. 32/50
    28/34
    905 is the more to right; the player 906 it is the most above; it's the player 907 is the one below. Such FOV it might be useful, per example, for picture one scenario attack or an
    concentration of players. With reference to Figures 9A and 9C, a second FOV 901 can be calculated so that it limits all moving players on the playing field 102. In this configuration, player 908 is in turn the leftmost player, and FOV includes a view of offensive zone 303 and neutral zone 302. Such a FOV can be useful, for example, for an escape play scenario.
    [079] The tracking camera that most players have is assigned to the tracking camera zone that is used in the process implemented in software to determine gameplay and select the current gameplay scenario. Data in the remaining zones can be ignored for the purpose of determining gameplay, in order to reduce the complexity of processing, for example.
    [080] To determine whether players are moving from one zone of interest (also called a tracking camera zone) to another, a variable is used to store the previous zone of interest. If the previous zone id is less than the current zone id then the players are moving to the right. If the previous zone id is greater than the current zone id then players move to the left. If the previous zone id is the same as the current zone id then players move within the same zone as tracking cameras. This technique is sufficient to detect transitions for players who move slowly, but it may not be enough for players who move quickly. Detecting moving players or groups of players
    Petition 870170035862, of 05/29/2017, p. 33/50
    29/34 can quickly require a predictive process.
    [081] To facilitate the detection of players moving from one area to another, position x left and right firing markers are used. If the group of players is moving (the x direction is not zero) within a zone, and if the group's x position falls with the marker and the associated zone border, and if the group's speed is greater than a limit , then the forecast is that the group is in transition into the new camera zone.
    [082] The current tracking camera zone and the predicted tracking camera zone can be used in the process to determine gameplay.
    [083] In some modalities, tracking cameras are not able to detect players who are not moving and therefore do not send mosaic data to the processing system for stopped players. The processing system must detect when a player is not moving by tracking the speed and direction of the
    players. When the velocity current one player if approaches a value limit  x of 0, the player is considered stopped at the positions, speed and direction of
    player are stored in a list of stopped players. During player tracking, if a new player is detected, his position is compared to players on the stopped list. If there is a match then the new player is a previously stopped player. The values of the players in the stopped list have a lifetime so that after time t the values of the players are deleted if they do not correspond to a player in motion.
    [084] The stopped players can be used in maneuver
    Petition 870170035862, of 05/29/2017, p. 34/50
    30/34 game restart where there are several players standing and a single judge or player moving to a new position. In a process based on pure priority movement, a judge or player moving will cause the process to generate horizontal movement, vertical movement and zoom values so that the transmitting camera follows the movement. With the present method, the process recognizes that players are stopped in a game restart circle, and selects a game restart scenario. In the game restart gameplay scenario, the broadcast camera is moved horizontally, moved vertically and zoomed in to focus on the game restart area where a player or group of players is located. Even if there is the movement of a single player, the process will maintain the horizontal movement and vertical movement values of the current broadcast camera instead of following the movement of the single player. This state is maintained until the game restart ends and the process selects a different gameplay scenario.
    [085] A breakaway gameplay scenario can be identified by analyzing activity such as a combination of player group dynamics (speed and position) and single player behavior. With reference to Figure 7, the following sequence of events 700 must occur to predict a breakaway:
    - All players (the group) are in any of the tracking camera zones 1 or 4 (terminal zone) for a certain period of time t, 702;
    - A player (breakaway player) moves to another zone (2 or 3) at a higher speed than the group of players
    Petition 870170035862, of 05/29/2017, p. 35/50
    31/34
    704 and a distance greater than d from group 706; and
    - The group of players must move within the terminal zone in the same direction as the breakaway player or move to another zone, 708.
    [086] When examining player and group activity, the system can analyze player and group dynamics in a single zone of interest that contains the largest number of players. In scenarios where the game is changing rapidly, the sequence may include step 710 of predicting whether a player or group is moving to zone 2 or 3. If so, the system can analyze activity within both the predicted zone and the zone that currently contains the majority of players.
    [087] The gameplay scenario can change frequently during a game. The selection of the gameplay scenario may therefore involve a continuous analysis of the gameplay activity in order to select the appropriate scenario. The selection may involve a decision-making tree that identifies patterns or a sequence of events indicative of a specific scenario. Once the scenario is selected, the PTZ of the broadcast camera can be adjusted to better represent the current gameplay within the frame of the transmissible video.
    [088] With reference to Figure 8, a method for selecting a current gameplay scenario uses the parameters described above to identify a specific gameplay scenario and adjust the PTZ values of the broadcast camera accordingly. If the players are moving 802, and the players are leaving an 804 restart area, the selected scenario is a restart of
    Petition 870170035862, of 05/29/2017, p. 36/50
    32/34 game finished. The camera moves away from the 806 game restart area to show more of the ice. If players are on the move 802, but are not leaving an 804 game restart, the selected scenario is standard game. The camera is moved horizontally, moved vertically and enlarged 822 to follow the movement of the players. If the sequence of events described in Figure 7 occurs, the selected scenario is a breakaway 818. The breakaway player and zone are found 820 and the camera is moved horizontally, moved vertically and zoomed in 822 to focus on the breakaway player.
    [089] If the players are not moving 802 at the beginning of the process, but the players are in the 810 game restart area, the selected scenario is a game restart. The transmission camera is moved horizontally, moved vertically and enlarged to focus the game restart circle. If players are not in the 810 game restart circle, the camera is moved vertically and zoomed in to properly follow the action 814.
    [090] Still referring to Figure 8, once the PTZ values have been calculated, the values are transmitted from the computer to the transmission camera using an API 808.
    [091] The method may also involve the additional step of composing the scoreboard video on top of the video recorded by the broadcast camera. With reference to Figure 10, a composite video 1000 may include video images from the scoreboard 1004 overlaid on a portion of the video captured by the broadcast camera 1002. In this way a viewer watching the video will be able to always see the score of the game. At
    Petition 870170035862, of 05/29/2017, p. 37/50
    33/34
    modality illustrated, the images of the camera in scoreboard are composed directly on the images gives camera in streaming Preferably, the video in scoreboard is
    positioned so that it does not visually interfere with the video from the broadcast camera. Scoreboard images may be faded or temporarily hidden if it is determined that the scoreboard video may interfere with the video on the broadcast camera, for example if it covers a player. In one mode, the scoreboard video can be processed in order to extract the information it displays. This information can be used to automatically generate a personalized graphic representative of the scoreboard to overlay the video from the broadcast camera. Once composed, the resulting video can be transmitted to client devices over the internet or any network. It can be transmitted in the form of live images of the game, or saved for later transmission.
    [092] The steps listed are just a few possible embodiments of the invention described herein. Other modalities allow some of the steps to be performed in a different order, to be performed on different components, or for several steps to be performed on a single component. For example, mosaic data can be generated in the same computational unit that determines gameplay, instead of being generated in a separate computational module in the tracking camera unit.
    [093] According to other possible modalities, when using the method to track sports other than hockey, additional steps can be involved to identify and select different specific gameplay scenarios
    Petition 870170035862, of 05/29/2017, p. 38/50
    34/34 to private sport. Also, according to other modalities, the limits and dimensions of pixels used in the different processes can vary well and need to be calibrated according to the needs of the system. The specific values described in the modality presented above are for example purposes only and should not be considered as limiting the scope of the invention in any way.
    权利要求:
    Claims (21)
    [1]
    1. Method to automatically capture transmissible video of a sporting event inside a stadium, the CHARACTERIZED method for understanding the steps of:
    - divide the stadium into a plurality of zones of interest;
    - optically track individual players in the stadium by determining the respective positions of individual players within the zones of interest using at least one fixed tracking camera;
    - based on the respective positions of the individual players, perform a cluster analysis to identify groups of players within the zones of interest;
    - determine a particular zone of the zones of interest that contains the largest number of individual players;
    - analyze the activity of individual players and groups of players within the given zone of the zones of interest to select a current gameplay scenario corresponding to that activity, the current gameplay scenario being selected from a predetermined set of gameplay scenarios; and
    - Based on the current gameplay scenario, adjust the control of horizontal movement, vertical movement and zoom of a transmission camera in real time, thus changing a field of view of the transmission camera and capturing the transmissible video of the sporting event.
    [2]
    2. Method, according to claim 1, CHARACTERIZED by the fact that the current gameplay scenario is selected based on a position, speed or direction
    Petition 870170035854, of 05/29/2017, p. 6/15
    2/9 current of at least one of the individual players or at least one of the groups of players in relation to the determined zone of the zones of interest.
    [3]
    3. Method, according to claim 1 or 2, CHARACTERIZED because it also includes identifying individual players or groups of players that transit between adjacent zones of interest, and where the current gameplay scenario is selected based on a position, speed or direction current status of individual players or groups of players who move between adjacent zones.
    [4]
    4. Method, according to claim 3, CHARACTERIZED by also comprising predicting a transition of individual players or groups of players between first and second adjacent zones, and where the activity of individual players or groups of players is analyzed within both the first and second zones in order to select the current gameplay scenario.
    [5]
    5. Method, according to any one of claims 1 to 4, CHARACTERIZED by the fact that the optical tracking of individual players comprises:
    - extract foreground mosaics from frames of tracking video captured by at least one fixed tracking camera;
    - determine locations of foreground mosaics based on their positions within the frames of the tracking video;
    - correlate each of the foreground mosaics to a respective player of the individual players; and
    - maintain a list of tracked players, each player tracked on the list being associated with a number
    Petition 870170035854, of 05/29/2017, p. 7/15
    3/9 predetermined of the player's last known locations.
    [6]
    6. Method, according to claim 5, CHARACTERIZED in that the correlation of foreground mosaics comprises matching the location of a foreground mosaic to a player on the list that has a last known location closest to the foreground mosaic, and if a match is found, update the list so that the last most recent known location associated with the player matches the location of the foreground mosaic, and if no match is found, add a new tracked player to the list with a last location most recent known value corresponding to the location of the foreground mosaic.
    [7]
    7. Method, according to claim 5 or 6, CHARACTERIZED by the fact that the extraction of the foreground mosaics from the tracking video frames captured by at least one fixed tracking camera comprises subtracting sequential frames from the tracking video to obtain an image difference, and extract the foreground mosaics from the difference image by:
    - applying a limit to the difference image to obtain a binary image of the foreground mosaics on a background;
    - obscuring the binary image to remove noise; and
    - execution of a contour algorithm in the binary image to extract the mosaics.
    [8]
    8. Method, according to claim 7, CHARACTERIZED for also comprising identifying mosaics in
    Petition 870170035854, of 05/29/2017, p. 8/15
    4/9 movement corresponding to moving players and stationary mosaics corresponding to stationary players, where moving mosaics are extracted from the difference image and stationary mosaics are extracted from a single frame of the tracking video.
    [9]
    9. Method, according to any of claims 1 to 8, CHARACTERIZED by the fact that the selection of the current gameplay scenario comprises matching the activity of individual players and groups of players to a predetermined pattern associated with a specific gameplay scenario in predetermined set of gameplay scenarios.
    [10]
    10. Method according to any one of claims 1 to 9, CHARACTERIZED by the fact that the stadium is divided so that the zones of interest correspond to zones of a playing field within the stadium.
    [11]
    11. Method, according to any one of claims 1 to 10, CHARACTERIZED by further comprising calibrating at least one fixed tracking camera by mapping a tracking region on a playing field within the stadium, the tracking region being a region within which the tracking video is used to optically track individual players, and outside of which the tracking video is ignored.
    [12]
    12. Method according to any one of claims 1 to 11, CHARACTERIZED by the fact that individual players are tracked using a plurality of fixed tracking cameras, and by also understanding to normalize tracking data through the plurality of tracking cameras, and merging tracking data
    Petition 870170035854, of 05/29/2017, p. 9/15
    5/9 of all the plurality of tracking cameras for the purpose of tracking individual players globally within the stadium.
    [13]
    13. Method according to any one of claims 1 to 12, CHARACTERIZED because it also comprises performing a cluster analysis on the position of individual players to identify a global group of players within the stadium, and selecting the gameplay scenario based on a current position, speed or direction of the global group.
    [14]
    14. Method, according to any one of claims 1 to 13, CHARACTERIZED by including still capturing the video of the scoreboard and combining said video of the scoreboard on the transmissible video, the video of the scoreboard comprising video of a scoreboard inside the stadium.
    [15]
    15. Method according to any one of claims 1 to 14, CHARACTERIZED by the fact that the horizontal movement control, vertical movement and zoom of the transmission camera is adjusted so that the field of view is limited to contain a leftmost player , a player farther to the right, a player higher and a player lower within the specified zone of the zones of interest or within the stadium.
    [16]
    16. Method according to any one of claims 1 to 15, CHARACTERIZED in that it also comprises transmitting the transmissible video as live images over a network.
    [17]
    17. System to automatically capture transmissible video of a sporting event inside a stadium, the system FEATURED for understanding:
    Petition 870170035854, of 05/29/2017, p. 10/15
    6/9 at least one fixed tracking camera configured to capture tracking video of players in the stadium;
    a broadcast camera configured to capture the broadcast video of the sporting event, the broadcast camera being remotely controlled to adjust its horizontal movement, vertical movement and zoom control; and a processing system operatively coupled to at least one tracking camera and the transmission camera, the processing system configured for:
    - divide the stadium into a plurality of zones of interest;
    - optically track individual players in the stadium by determining the respective positions of individual players within the zones of interest using at least one fixed tracking camera;
    - based on the respective positions of the individual players, perform a cluster analysis to identify groups of players within the zones of interest;
    - determine a particular zone of the zones of interest that contains the largest number of individual players;
    - analyze the activity of individual players and groups of players within the given zone of the zones of interest to select a current gameplay scenario corresponding to that activity, the current gameplay scenario being selected from a predetermined set of gameplay scenarios; and
    - based on the current gameplay scenario, adjust the horizontal movement, vertical movement and zoom control
    Petition 870170035854, of 05/29/2017, p. 11/15
    Ί / 9 of the transmission camera in real time, thus changing a field of view of the transmission camera and capturing the transmissible video of the sporting event.
    [18]
    18. System, according to claim 17, CHARACTERIZED by the fact that at least one tracking camera is a plurality of tracking cameras, and by the fact that each of the plurality of tracking cameras is designed to capture tracking video from a determined zone of the plurality of zones of interest in the stadium.
    [19]
    19. System, according to claim 17 or 18, CHARACTERIZED by the fact that at least one tracking camera comprises an optical tracking module, the optical tracking module being configured to:
    - extract foreground mosaics from tracking video frames captured by at least one fixed tracking camera;
    - determine locations of foreground mosaics based on their positions within the frames of the tracking video; and
    - transmit coordinates of the locations of the foreground mosaics to the processing system, said coordinates being transmitted without also transmitting the tracking video to the processing system.
    [20]
    20. System, according to any one of claims 17 to 19, CHARACTERIZED by also comprising a scoreboard camera operatively coupled to the processing system, the scoreboard camera being configured to capture scoreboard video inside the stadium, and where the processing is further configured to combine the video
    Petition 870170035854, of 05/29/2017, p. 12/15
    8/9 score with transmissible video.
    [21]
    21. Machine-readable non-transient medium CHARACTERIZED for executing one or more sequences of instructions that, when executed by one or more processors operatively coupled to at least one tracking camera and a remotely controllable transmission camera in a stadium during an event sports, causes the one or more processors to perform the steps of:
    - divide the stadium into a plurality of zones of interest;
    - optically track individual players in the stadium by determining the respective positions of individual players within the zones of interest using at least one fixed tracking camera;
    - based on the respective positions of the individual players, perform a cluster analysis to identify groups of players within the zones of interest;
    - determine a particular zone of the zones of interest that contains the largest number of individual players;
    - analyze the activity of individual players and groups of players within the given zone of the zones of interest to select a current gameplay scenario corresponding to that activity, the current gameplay scenario being selected from a predetermined set of gameplay scenarios; and
    - Based on the current gameplay scenario, adjust the horizontal movement control, vertical movement and zoom of the transmission camera in real time, thereby changing a field of view of the transmission camera and capturing the
    Petition 870170035854, of 05/29/2017, p. 13/15
    9/9 transmissible video of the sporting event.
    类似技术:
    公开号 | 公开日 | 专利标题
    BR112017007408A2|2019-11-19|system and method for optical tracking of players in sports stadiums
    JP6719465B2|2020-07-08|System and method for displaying wind characteristics and effects in broadcast
    CA2798298C|2016-08-23|Systems and methods for video processing
    ES2845933T3|2021-07-28|Automatic television production method and system
    US20150297949A1|2015-10-22|Automatic sports broadcasting system
    US9094615B2|2015-07-28|Automatic event videoing, tracking and content generation
    ES2805800T3|2021-02-15|Method and system to produce a video production
    JP2008520016A|2008-06-12|Image-based motion tracking
    US20200404174A1|2020-12-24|Computer-implemented method for automated detection of a moving area of interest in a video stream of field sports with a common object of interest
    US20170366867A1|2017-12-21|Systems and methods for displaying thermographic characteristics within a broadcast
    US20180225852A1|2018-08-09|Apparatus and method for generating best-view image centered on object of interest in multiple camera images
    CN110049345A|2019-07-23|A kind of multiple video strems director method and instructor in broadcasting's processing system
    JP2020095717A|2020-06-18|Method, system and apparatus for capture of image data for free viewpoint video
    US20170366866A1|2017-12-21|Systems and methods for displaying wind characteristics and effects within a broadcast
    CA2901913C|2016-04-12|System and method for optical player tracking in sports venues
    WO2020230751A1|2020-11-19|Disposition determination device, system, disposition determination method, and program
    JP2020191618A|2020-11-26|Arrangement determination device, system, arrangement determination method, and program
    US11188759B2|2021-11-30|System and method for automated video processing of an input video signal using tracking of a single moveable bilaterally-targeted game-object
    KR20220017217A|2022-02-11|System for filming and analyzing of sport play
    CN111970434A|2020-11-20|Multi-camera multi-target athlete tracking shooting video generation system and method
    同族专利:
    公开号 | 公开日
    AU2015330646A1|2017-06-01|
    JP2017531979A|2017-10-26|
    WO2016054729A1|2016-04-14|
    CN107005654A|2017-08-01|
    KR20170086493A|2017-07-26|
    US9468850B2|2016-10-18|
    US20160101358A1|2016-04-14|
    MX2017004722A|2017-11-30|
    RU2017116214A|2018-11-12|
    EP3205112A1|2017-08-16|
    US9744457B2|2017-08-29|
    US20170001118A1|2017-01-05|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US6124862A|1997-06-13|2000-09-26|Anivision, Inc.|Method and apparatus for generating virtual views of sporting events|
    US6270433B1|1997-12-15|2001-08-07|Toy Builders|Player position detection system|
    JP3985369B2|1998-08-21|2007-10-03|株式会社セガ|Game screen display control method, character movement control method, game machine, and recording medium recording program|
    US20020030742A1|1998-11-20|2002-03-14|Aman James A.|Employing electomagnetic by-product radiation for object tracking|
    US6707487B1|1998-11-20|2004-03-16|In The Play, Inc.|Method for representing real-time motion|
    US8965898B2|1998-11-20|2015-02-24|Intheplay, Inc.|Optimizations for live event, real-time, 3D object tracking|
    AU2002363942A1|2001-11-20|2003-06-10|James A. Aman|Optimizations for live event, real-time, 3d object traking|
    US6567116B1|1998-11-20|2003-05-20|James A. Aman|Multiple object tracking system|
    US9555310B2|1998-11-20|2017-01-31|Maxx Holdings, Inc.|Sports scorekeeping system with integrated scoreboard and automatic entertainment system|
    US7483049B2|1998-11-20|2009-01-27|Aman James A|Optimizations for live event, real-time, 3D object tracking|
    CA2328048E|2000-12-11|2013-04-09|Hockeyline Inc.|Electronic scorekeeping device and system therefor|
    GB2379571A|2001-09-11|2003-03-12|Eitan Feldbau|Determining the Position of Players on a Sports Field|
    US6950123B2|2002-03-22|2005-09-27|Intel Corporation|Method for simultaneous visual tracking of multiple bodies in a closed structured environment|
    KR20050061607A|2002-11-20|2005-06-22|가부시키가이샤 세가|Game image display control program, game device, and recording medium|
    US8614741B2|2003-03-31|2013-12-24|Alcatel Lucent|Method and apparatus for intelligent and automatic sensor control using multimedia database system|
    JP3656118B2|2003-09-25|2005-06-08|株式会社コナミコンピュータエンタテインメント東京|GAME DEVICE, COMPUTER CONTROL METHOD, AND PROGRAM|
    US6867116B1|2003-11-10|2005-03-15|Macronix International Co., Ltd.|Fabrication method of sub-resolution pitch for integrated circuits|
    WO2005099423A2|2004-04-16|2005-10-27|Aman James A|Automatic event videoing, tracking and content generation system|
    US8723956B2|2004-08-30|2014-05-13|Trace Optic Technologies Pty Ltd|Method and apparatus of camera control|
    US9327191B2|2006-05-08|2016-05-03|Nintendo Co., Ltd.|Method and apparatus for enhanced virtual camera control within 3D video games or other computer graphics presentations providing intelligent automatic 3D-assist for third person viewpoints|
    US8611723B2|2006-09-06|2013-12-17|James Andrew Aman|System for relating scoreboard information with event video|
    WO2008033338A2|2006-09-11|2008-03-20|Aman James A|System and methods for translating sports tracking data into statistics and performance measurements|
    US8077981B2|2007-07-27|2011-12-13|Sportvision, Inc.|Providing virtual inserts using image tracking with camera and position sensors|
    JP5390115B2|2008-03-31|2014-01-15|株式会社バンダイナムコゲームス|Program, game system|
    US8339456B2|2008-05-15|2012-12-25|Sri International|Apparatus for intelligent and autonomous video content generation and streaming|
    US9185361B2|2008-07-29|2015-11-10|Gerald Curry|Camera-based tracking and position determination for sporting events using event information and intelligence data extracted in real-time from position information|
    US20110173235A1|2008-09-15|2011-07-14|Aman James A|Session automated recording together with rules based indexing, analysis and expression of content|
    US8622795B2|2008-12-04|2014-01-07|Home Box Office, Inc.|System and method for gathering and analyzing objective motion data|
    US8698898B2|2008-12-11|2014-04-15|Lucasfilm Entertainment Company Ltd.|Controlling robotic motion of camera|
    US9269154B2|2009-01-13|2016-02-23|Futurewei Technologies, Inc.|Method and system for image processing to classify an object in an image|
    US9462030B2|2009-03-04|2016-10-04|Jacquelynn R. Lueth|System and method for providing a real-time three-dimensional digital impact virtual audience|
    JP5299018B2|2009-03-26|2013-09-25|ソニー株式会社|Information processing apparatus, content processing method, and program|
    US8786415B2|2010-02-24|2014-07-22|Sportvision, Inc.|Tracking system using proximity and/or presence|
    US9600760B2|2010-03-30|2017-03-21|Disney Enterprises, Inc.|System and method for utilizing motion fields to predict evolution in dynamic scenes|
    US8456518B2|2010-03-31|2013-06-04|James Cameron & Vincent Pace|Stereoscopic camera with automatic obstruction removal|
    US8170413B2|2010-03-31|2012-05-01|James Cameron|Single seat-kill camera system|
    EP2618897A4|2010-07-08|2016-09-07|Intheplay Inc|Sports scorekeeping system with integrated scoreboard and automatic entertainment system|
    US9626786B1|2010-07-19|2017-04-18|Lucasfilm Entertainment Company Ltd.|Virtual-scene control device|
    US9055226B2|2010-08-31|2015-06-09|Cast Group Of Companies Inc.|System and method for controlling fixtures based on tracking data|
    US8854594B2|2010-08-31|2014-10-07|Cast Group Of Companies Inc.|System and method for tracking|
    JP5614211B2|2010-09-30|2014-10-29|株式会社セガ|Image processing program and computer-readable recording medium|
    US9375628B2|2010-11-19|2016-06-28|Isolynx, Llc|Associative object tracking systems and methods|
    US20120142429A1|2010-12-03|2012-06-07|Muller Marcus S|Collaborative electronic game play employing player classification and aggregation|
    US8659663B2|2010-12-22|2014-02-25|Sportvision, Inc.|Video tracking of baseball players to determine the start and end of a half-inning|
    JP5725953B2|2011-04-13|2015-05-27|キヤノン株式会社|Imaging apparatus, control method therefor, and information processing apparatus|
    US9298986B2|2011-12-09|2016-03-29|Gameonstream Inc.|Systems and methods for video processing|
    US8704904B2|2011-12-23|2014-04-22|H4 Engineering, Inc.|Portable system for high quality video recording|
    US9363488B2|2012-01-06|2016-06-07|Nokia Technologies Oy|Methods, apparatuses and computer program products for analyzing crowd source sensed data to determine information related to media content of media capturing devices|
    US9202526B2|2012-05-14|2015-12-01|Sstatzz Oy|System and method for viewing videos and statistics of sports events|
    US20130300832A1|2012-05-14|2013-11-14|Sstatzz Oy|System and method for automatic video filming and broadcasting of sports events|
    WO2014008504A1|2012-07-06|2014-01-09|H4 Engineering, Inc.|A remotely controlled automatic camera tracking system|
    US8968100B2|2013-02-14|2015-03-03|Sstatzz Oy|Sports training apparatus and method|
    CA3104723A1|2013-04-29|2014-10-29|Intelliview Technologies Inc.|Object detection|
    US9180357B2|2013-06-06|2015-11-10|Zih Corp.|Multiple antenna interference rejection in ultra-wideband real time locating systems|
    US9319641B2|2013-07-24|2016-04-19|The Boeing Company|Controlling movement of a camera to autonomously track a mobile object|
    US20150189243A1|2013-12-27|2015-07-02|Telemetrio LLC|Automated video production system|
    CN104038737A|2014-05-30|2014-09-10|西安交通大学|Double-camera system and method for actively acquiring high-resolution image of interested target|
    WO2016007962A1|2014-07-11|2016-01-14|ProSports Technologies, LLC|Camera feed distribution from event venue virtual seat cameras|
    US9498678B2|2014-07-11|2016-11-22|ProSports Technologies, LLC|Ball tracker camera|
    US9571903B2|2014-07-11|2017-02-14|ProSports Technologies, LLC|Ball tracker snippets|
    US9398213B1|2014-07-11|2016-07-19|ProSports Technologies, LLC|Smart field goal detector|US10456209B2|2010-10-13|2019-10-29|Gholam A. Peyman|Remote laser treatment system with dynamic imaging|
    WO2019183235A1|2018-03-21|2019-09-26|Second Spectrum, Inc.|Methods and systems of spatiotemporal pattern recognition for video content development|
    US10769446B2|2014-02-28|2020-09-08|Second Spectrum, Inc.|Methods and systems of combining video content with one or more augmentations|
    US10521671B2|2014-02-28|2019-12-31|Second Spectrum, Inc.|Methods and systems of spatiotemporal pattern recognition for video content development|
    US11120271B2|2014-02-28|2021-09-14|Second Spectrum, Inc.|Data processing systems and methods for enhanced augmentation of interactive video content|
    JP2016046642A|2014-08-21|2016-04-04|キヤノン株式会社|Information processing system, information processing method, and program|
    NO20161098A1|2016-07-01|2018-01-02|Teameye As|Camera system for filming sports venues|
    US10530972B2|2016-09-21|2020-01-07|Htc Corporation|Control method for optical tracking system|
    JP6821398B2|2016-11-09|2021-01-27|キヤノン株式会社|Image processing equipment, image processing methods and programs|
    US10269133B2|2017-01-03|2019-04-23|Qualcomm Incorporated|Capturing images of a game by an unmanned autonomous vehicle|
    JP6965975B2|2017-03-10|2021-11-10|株式会社Ｊｖｃケンウッド|Video processing equipment, multi-camera system, video processing method, and video processing program|
    US10674094B2|2017-06-16|2020-06-02|Skycam, Llc|Infinite virtual pan tilt zoom camera|
    WO2019014521A1|2017-07-13|2019-01-17|Peyman Gholam A|Dynamic image recognition system for security and telemedicine|
    WO2019025833A1|2017-08-02|2019-02-07|Playgineering Systems, Sia|A system and a method for automated filming|
    US10885310B2|2017-08-17|2021-01-05|Jpmorgan Chase Bank, N.A.|Sytems and methods for object recognition and association with an identity|
    JP6972804B2|2017-09-11|2021-11-24|株式会社Ｊｖｃケンウッド|Chat terminal device, chat system, chat display method, and chat display program|
    GB2581746A|2017-10-22|2020-08-26|Martin Todd|System and method for image recognition registration of an athlete in a sporting event|
    US10896357B1|2017-12-29|2021-01-19|Automation Anywhere, Inc.|Automatic key/value pair extraction from document images using deep learning|
    CN108211319B|2018-01-30|2019-06-18|平顶山学院|Multifunctional intellectual scoring table for sport track and field|
    RU2683499C1|2018-03-15|2019-03-28|Антон Владимирович Роженков|System for automatic creation of scenario video clip with preset object or group of objects presence in frame|
    US10769427B1|2018-04-19|2020-09-08|Automation Anywhere, Inc.|Detection and definition of virtual objects in remote screens|
    WO2019229748A1|2018-05-28|2019-12-05|United Perfectum Ltd.|Golf game video analytic system|
    US10904076B2|2018-05-30|2021-01-26|International Business Machines Corporation|Directing functioning of an object based on its association to another object in an environment|
    US11172248B2|2019-01-22|2021-11-09|Tempus Ex Machina, Inc.|Systems and methods for customizing and compositing a video feed at a client device|
    US11140328B2|2019-01-22|2021-10-05|Tempus Ex Machina, Inc.|Systems and methods for partitioning a video feed to segment live player activity|
    US10918945B2|2019-03-15|2021-02-16|Sony Interactive Entertainment Inc.|Methods and systems for spectating characters in follow-mode for virtual reality views|
    US11058950B2|2019-03-15|2021-07-13|Sony Interactive Entertainment Inc.|Methods and systems for spectating characters in virtual reality views|
    CN110213611A|2019-06-25|2019-09-06|宫珉|A kind of ball competition field camera shooting implementation method based on artificial intelligence Visual identification technology|
    CN110300288A|2019-07-29|2019-10-01|富咖科技有限公司|It is a kind of for touching the video automatic collection and synthetic method played in zoo|
    CN112637612B|2019-09-24|2021-11-23|广州虎牙科技有限公司|Live broadcast platform and interactive video processing method thereof|
    US11113535B2|2019-11-08|2021-09-07|Second Spectrum, Inc.|Determining tactical relevance and similarity of video sequences|
    KR102112517B1|2020-03-06|2020-06-05|모바일센 주식회사|Unmanned sports relay service method through real time video analysis and video editing and apparatus for same|
    CN111541846A|2020-05-07|2020-08-14|元动未来科技有限公司|Automatic ice kettle image following and shooting system|
    US20210379492A1|2020-06-09|2021-12-09|International Business Machines Corporation|Real-world activity simulation augmentation with real-world data of the activity|
    法律状态:
    2019-11-26| B11A| Dismissal acc. art.33 of ipl - examination not requested within 36 months of filing|
    2020-02-11| B11Y| Definitive dismissal - extension of time limit for request of examination expired [chapter 11.1.1 patent gazette]|
    2021-10-13| B350| Update of information on the portal [chapter 15.35 patent gazette]|
    优先权:
    申请号 | 申请日 | 专利标题
    US201462062457P| true| 2014-10-10|2014-10-10|
    PCT/CA2015/050409|WO2016054729A1|2014-10-10|2015-05-08|System and method for optical player tracking in sports venues|
    [返回顶部]