CN112200891A - System and method for reconstructing video three-dimensional animation scene of football match - Google Patents
System and method for reconstructing video three-dimensional animation scene of football match Download PDFInfo
- Publication number
- CN112200891A CN112200891A CN202010804502.5A CN202010804502A CN112200891A CN 112200891 A CN112200891 A CN 112200891A CN 202010804502 A CN202010804502 A CN 202010804502A CN 112200891 A CN112200891 A CN 112200891A
- Authority
- CN
- China
- Prior art keywords
- football
- player
- information
- judging
- video frame
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T13/00—Animation
- G06T13/20—3D [Three Dimensional] animation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/49—Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Abstract
The invention discloses a three-dimensional animation scene reconstruction system and a method aiming at a football match video, wherein the system comprises: the football match video frame sequence input module is used for inputting a video frame sequence of a football match; the court information extraction module is used for extracting the court information on each video frame; the player identification tracking module is used for automatically identifying and tracking players on each video frame; the football identification tracking module is used for automatically identifying and tracking the football motion track on each video frame; the three-dimensional animation scene reconstruction module is used for providing the designer with a three-dimensional animation scene for reconstructing a football match according to the extracted court information, the identified and tracked player information and the football information; and the player action analysis module is used for automatically analyzing the action of the players according to the player information and the football information. The reconstructed three-dimensional animation scene allows the spectators to watch the game at any visual angle, and the action behaviors of each player on the game field can be automatically analyzed.
Description
Technical Field
The invention relates to the technical field of football match video three-dimensional scene and animation construction, in particular to a three-dimensional animation scene reconstruction system and method aiming at a football match video.
Background
The football match video that video recording equipment recorded only can be with the broadcast of video recording visual angle, and the spectator can't watch the football match video of recording with arbitrary visual angle, consequently, in some special demand scenes, for example football coach need show through watching video analysis sportsman in the later stage, carry out analysis etc. to the other party's team tactics, because the football match video of recording can't watch with arbitrary visual angle, show for football coach analysis sportsman, the obstacle has been brought to other party's team tactics etc. of analysis.
In addition, the recorded football game video does not have the function of automatically analyzing the action behaviors of the players, and the coach evaluates the game states of the players by artificially analyzing the action behaviors of each player during the live period, such as whether the football player has a shooting behavior, whether the football player has a snapping behavior, whether the football player has a pass behavior and the like. The artificial analysis method for the action and the behavior of the players is quite tired, analysis omission easily occurs, and objective evaluation on the match state of each player cannot be carried out.
Disclosure of Invention
The invention aims to provide a three-dimensional animation scene reconstruction system and a three-dimensional animation scene reconstruction method aiming at a football match video, so as to solve the technical problems.
In order to achieve the purpose, the invention adopts the following technical scheme:
provided is a three-dimensional animation scene reconstruction system for a football match video, which comprises:
the football match video frame sequence input module is used for inputting a video frame sequence of a football match;
the court information extraction module is connected with the football match video frame sequence input module and is used for extracting court information representing court characteristics on each video frame;
the player identification tracking module is connected with the football match video frame sequence input module and is used for automatically identifying and tracking players on each video frame;
the football identification tracking module is connected with the football match video frame sequence input module and is used for automatically identifying and tracking the football motion track on each video frame;
the three-dimensional animation scene reconstruction module is respectively connected with the court information extraction module, the player identification tracking module and the football identification tracking module and is used for providing designers with three-dimensional animation scenes for reconstructing a football match according to the extracted court information, the identified and tracked player information and football information;
sportsman's action analysis module connects respectively sportsman discernment tracker with football discernment tracker for according to discernment, trail sportsman's information with football information analyzes sportsman's action, and output analysis result.
As a preferable aspect of the present invention, the player identification tracking module includes:
the player segmentation extraction unit is used for carrying out image segmentation on the video frame image so as to extract players on the video frame;
the team judgment unit is connected with the player division and extraction unit and used for judging the team to which the player belongs according to the extracted player characteristics to obtain a judgment result;
and the player motion tracking unit is connected with the player segmentation and extraction unit and used for realizing automatic tracking of the motion of each extracted player.
As a preferred aspect of the present invention, an image segmentation algorithm based on color information performs image segmentation on the video frame image to extract the players on the video frame.
As a preferred aspect of the present invention, the image segmentation algorithm based on color information is expressed by the following formula:
in the above formula, b (x, y) represents the position of the suspected region of the player on the video frame;
x represents the coordinate of the suspected area position in the x-axis direction;
y represents the coordinate of the suspected area position in the y-axis direction;
g (x, y) represents a color value of a green component of the suspected location of the area;
r (x, y) represents a color value of a red component of the location of the suspected area;
b (x, y) represents a color value of a blue component of the location of the suspected area;
Trand TbTwo color thresholds;
b (x, y) ═ 1 indicates that the player at the position of the suspected area is true;
b (x, y) ═ 0 indicates that the player at the position of the area where the plausibility exists is false.
As a preferred embodiment of the present invention, the color threshold T isr＝15。
As a preferred embodiment of the present invention, the color threshold T isb＝60。
As a preferred aspect of the present invention, the player action behavior analysis module specifically includes:
a ball-carrying behavior judging unit for judging whether the distance between the football and the same player in a continuous frame is always kept in a preset ball distance range according to the identified and tracked player information and the football information,
if so, judging that the current action behavior of the player is 'with the ball';
a kicking behavior judging unit for judging whether the football is always in the control range of the football player in a continuous frame before the current frame and the football leaves the control range of the football player in the current frame according to the identified and tracked information of the football player and the football information,
if so, judging that the current action behavior of the player is 'kicking the ball';
the kicking behavior judging unit specifically comprises:
a pass behavior judging subunit for judging whether the football far away from the kickers is got by the team players,
if yes, judging the kicking behavior of the player as 'pass';
a snap action judgment subunit for judging whether the football far away from the kicker is got by the opponent team player,
if yes, judging the action behavior of the opposite player as 'snap';
a shooting behavior judging subunit, which is used for judging whether the football far away from the kickers approaches the goal or not,
if yes, judging the kicking behavior of the player as shooting;
the attack and defense state judging unit is connected with the ball-carrying behavior judging unit and used for judging whether opposite team players exist in a preset distance range from the ball-carrying players or not according to the identified and tracked player information and football information after judging that the current action behavior of the players is 'ball-carrying',
if yes, the player with the ball is judged to be in an attack and defense state at present.
The invention also provides a three-dimensional animation scene reconstruction method for the football match video, which is realized by applying the three-dimensional animation scene reconstruction system, and the method comprises the following steps:
inputting a video frame sequence of a football game;
extracting court information on each video frame;
automatically identifying and tracking the players on each video frame;
identifying and tracking the football track on each video frame;
constructing a static three-dimensional scene of the football match according to the court information extracted from the first frame and the player position information and the football position information identified from the first frame;
mapping the tracked player motion information and the tracked football motion trail to the static three-dimensional scene, and reconstructing the three-dimensional animation scene of the football match.
The invention also provides a player action behavior recognition method which is realized by applying the three-dimensional animation scene reconstruction system, and the method comprises the following steps:
inputting a video frame sequence of a football game;
extracting court information on each video frame;
automatically identifying and tracking the players on each video frame;
automatically identifying and tracking the football motion track on each video frame;
analyzing the action of the player according to the identified and tracked information of the player and the football information, and outputting an analysis result.
As a preferred scheme of the invention, the specific method for analyzing the action of the player comprises the following steps:
judging whether the distance between the football and the same player in a continuous frame is always kept in a preset ball distance range according to the identified and tracked player information and the football information,
if so, judging that the current action behavior of the player is 'with the ball';
according to the identified and tracked player information and football information, judging whether the football is always in the control range of the players in a continuous frame before the current frame and the football leaves the control range of the players in the current frame,
if so, judging that the current action behavior of the player is 'kicking the ball';
judging whether opponent team players exist in a preset distance range from the player with the ball according to the identified and tracked player information and the football information,
if yes, judging that the player with the ball is currently in an attack and defense state;
for the ball kicking behavior of the players, whether the football far away from the ball kicker is got by other players is further judged,
if yes, judging that the football player is a team member or an opponent team member, if the football player obtains the ball, judging that the action behavior of the kickball player is 'pass ball', and if the football player obtains the ball, judging that the current action behavior of the opponent player is 'snap off';
if not, judging whether the football far away from the kickers approaches to the goal, and if so, judging that the kicking behavior of the kickers is 'shooting'.
The invention extracts the court features, the player features and the football features based on the color difference values of lawns, players, football, goals, referees and the like on the video frame image of the competition field, automatically identifies and tracks the players and the football on the subsequent frame based on the player features and the football features extracted from the first frame, reconstructs the three-dimensional animation scene of the football match according to the extracted court information, the identified and tracked player information and football information, and allows audiences to watch the match at any visual angle. In addition, the invention can automatically analyze the action of each player on the playing field according to the identified and tracked information of the players and the football information, and the analysis result can be used as a basis for quantitatively evaluating the performance of the players.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below. It is obvious that the drawings described below are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 is a schematic structural diagram of a system for reconstructing a three-dimensional animation scene for a soccer game video according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of the player identification tracking module in the three-dimensional animated scene reconstruction system;
FIG. 3 is a schematic structural diagram of the player action behavior analysis module in the three-dimensional animation scene reconstruction system;
fig. 4 is a schematic structural diagram of the kicking behavior determination unit in the player action behavior analysis module;
FIG. 5 is a method step diagram of a method for reconstructing a three-dimensional animation scene for a football game video according to an embodiment of the present invention;
fig. 6 is a method step diagram of a player action behavior recognition method according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
Wherein the showings are for the purpose of illustration only and are shown by way of illustration only and not in actual form, and are not to be construed as limiting the present patent; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "inner", "outer", etc. are used for indicating the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not indicated or implied that the referred device or element must have a specific orientation, be constructed in a specific orientation and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and are not to be construed as limitations of the present patent, and the specific meanings of the terms may be understood by those skilled in the art according to specific situations.
In the description of the present invention, unless otherwise explicitly specified or limited, the term "connected" or the like, if appearing to indicate a connection relationship between the components, is to be understood broadly, for example, as being fixed or detachable or integral; can be mechanically or electrically connected; they may be directly connected or indirectly connected through intervening media, or may be connected through one or more other components or may be in an interactive relationship with one another. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The three-dimensional animation scene reconstruction system for a football match video provided by an embodiment of the present invention, as shown in fig. 1, includes:
the football match video frame sequence input module 1 is used for inputting a video frame sequence of a football match video clip;
the court information extraction module 2 is connected with the football match video frame sequence input module 1 and is used for extracting the court information on each video frame;
the player identification tracking module 3 is connected with the football match video frame sequence input module 1 and is used for automatically identifying and tracking players on each video frame;
the football identification tracking module 4 is connected with the football match video frame sequence input module 1 and is used for automatically identifying and tracking the football motion track on each video frame;
the three-dimensional animation scene reconstruction module 5 is respectively connected with the court information extraction module 2, the player identification tracking module 3 and the football identification tracking module 4 and is used for providing designers with three-dimensional animation scenes for reconstructing football matches according to the extracted court information, the identified and tracked player information and football information;
the player action analysis module 6 is respectively connected with the player identification tracking module 3 and the football identification tracking module 4, and is used for analyzing the action of players according to the identification and tracking of player information and football information and outputting the analysis result.
In this embodiment, it is preferable to perform image segmentation on the video frame image by using an image segmentation algorithm based on color information, so as to extract court information representing characteristics of a court and various types of characteristic information on the court such as a player, a football, a goal, a referee, and the like. Since there are many existing image segmentation algorithms based on color information that can be applied in the present invention, the specific procedures of the method for extracting the image features required on the video frame image are not described herein.
In this embodiment, to the player in the football match video identification track, need do the work of three aspects: segmentation extraction of players, identification of teams to which the players belong and tracking of the players. In order to realize the identification and tracking of the players, as shown in fig. 2, the player identification and tracking module 3 specifically includes:
a player segmentation extracting unit 31 for performing image segmentation on the video frame image to extract players on the video frame;
the team judgment unit 32 to which the player belongs is connected with the player segmentation and extraction unit 31 and used for judging the team to which the player belongs according to the extracted player characteristics to obtain a judgment result;
and the player movement tracking unit 33 is connected with the player segmentation extraction unit 31 and used for realizing automatic tracking of each extracted player movement.
The following explains the player characteristic segmentation and extraction process, the team judgment process of the players and the automatic tracking process of player movement:
firstly, according to the color characteristics of the football field, in the RGB color space, the color value of the green component of the pixels of the lawn area is greater than the color values of the red and blue components, the embodiment automatically segments the players from the video frame sequence by utilizing the statistical information of the color component difference. The invention extracts the characteristics of the players by the image segmentation algorithm based on the color information, and the image segmentation algorithm can be expressed by the following formula:
in the above formula, b (x, y) represents the suspected location of the player on the video frame;
x represents the coordinate of the position of the suspected area in the x-axis direction;
y represents the coordinate of the position of the suspected area in the y-axis direction;
g (x, y) represents a color value of a green component of the location of the suspected area;
r (x, y) represents the color value of the red component of the location of the suspected area;
b (x, y) represents a color value of a blue component of the location of the suspected area;
Trand TbTwo color thresholds;
b (x, y) ═ 1 indicates that the player at the position of the suspected area is true;
a b (x, y) — 0 indicates that the player at the position of the suspected area is false.
According to the invention, the value of the color threshold Tr is preferably 15 through repeated experimental calculation; the value of the color threshold value Tb is preferably 60. Under the constraint of the two color thresholds, the players who do not wear the green ball cover on the court with the green lawn as the background can be better segmented and extracted.
In order to improve the accuracy of extracting the player characteristics, before extracting the player characteristics, it is preferable to perform erosion-dilation processing on the marked area where the suspected player is located to filter out image noise in the area where the suspected player is located.
After the players are extracted by segmentation, the affiliated teams of the players need to be distinguished, which is important for the subsequent analysis of player action behaviors.
And finally, automatically tracking the movement track of the player by using the information of the player in the front and back frames, such as the movement speed, the position, the moving direction and the like of the player. There are many existing automatic tracking technologies for dynamic objects that can be applied in the present invention, so the process of automatic tracking for players and the process of automatic tracking for the motion trail of a soccer ball are not described here.
The invention focuses on the analysis of the action behaviors of the players, and the analysis result can quantitatively evaluate the performance of the players on the scene. Specifically, as shown in fig. 3, the player action behavior analysis module 6 specifically includes:
a determination unit 61 for determining whether the distance between the football and the same player in a continuous frame is always kept in a preset range of ball distance according to the identified and tracked information of the player (including the position, the speed, the acceleration and the direction of movement) and the information of the football (including the position and the movement track of the football),
if so, judging that the current action behavior of the player is 'with the ball';
a kicking behavior determination unit 62 for determining whether the football is always within the control range of the player (the control range can be represented by the distance between the football and the player, for example, when the distance between the football and the same player is always kept within 3 meters in the continuous frame range, it is represented that the football is always within the control range of the player) in a continuous frame before the current frame, and the football leaves the control range of the player in the current frame,
if so, judging that the current action behavior of the player is 'kicking the ball';
the attacking and defending state judging unit 63 is connected with the ball-carrying behavior judging unit 61 and is used for judging whether an opponent team member exists in a preset distance range from the ball-carrying player or not according to the identified and tracked player information and football information after judging that the current action behavior of the player is 'carrying ball',
if yes, the player with the ball is judged to be in an attack and defense state at present.
In order to further determine whether the "kicking" behavior of the player is "pass", "snap" or "shoot", it is more preferable that, as shown in fig. 4, the kicking behavior determination unit 62 specifically includes:
a pass behavior judging subunit 621 configured to judge whether or not the soccer ball far from the kicker is received by the teams,
if yes, judging the kicking behavior of the player as 'pass';
a snap action judging subunit 622 for judging whether the soccer ball far from the kicker is hit by the opponent team player,
if yes, judging the action behavior of the opposite player as 'snap';
a shooting behavior judging subunit 623 for judging whether the soccer far from the kicker approaches the goal,
if yes, the kicking behavior of the player is judged to be shooting.
After the action behaviors of each player on the competition field are analyzed, the times of different types of action behaviors made by each player can be counted to serve as a basis for quantitatively evaluating the performance of the player on the competition field.
The invention also provides a three-dimensional animation scene reconstruction method for the football match video, which is realized by applying the three-dimensional animation scene reconstruction system, and as shown in fig. 5, the three-dimensional animation scene reconstruction method comprises the following steps:
step S1, inputting a video frame sequence of the football game;
step S2, extracting court information on each video frame;
step S3, automatically identifying and tracking the player on each video frame;
step S4, identifying and tracking the football track on each video frame;
step S5, constructing a static three-dimensional scene of the football match according to the court information extracted from the first frame and the player position information and the football position information identified from the first frame;
and mapping the tracked player motion information and the tracked football motion track to a static three-dimensional scene to reconstruct a three-dimensional animation scene of the football match. There are many existing methods for mapping dynamic poses (including motion poses) of characters or objects onto target characters or object objects in a static three-dimensional scene to construct a three-dimensional animated scene, so the process of reconstructing a three-dimensional animated scene of a soccer game according to the present invention is not described herein.
The invention also provides a player action behavior identification method, which is realized by the three-dimensional animation scene reconstruction system, and as shown in fig. 6, the player action behavior identification method comprises the following steps:
step L1, inputting a sequence of video frames of a soccer game;
l2, extracting the court information on each video frame;
step L3, automatically identifying and tracking the player on each video frame;
l4, automatically identifying and tracking the football track on each video frame;
and L5, analyzing the action of the player according to the identified and tracked information of the player and the football information, and outputting the analysis result.
In step L5, the specific method for analyzing the action of the player is as follows:
judging whether the distance between the football and the same player in a continuous frame is always kept in a preset ball distance range according to the identified and tracked player information and the football information,
if so, judging that the current action behavior of the player is 'with the ball';
according to the identified and tracked player information and football information, judging whether the football is always in the control range of the players in a continuous frame before the current frame and the football leaves the control range of the players in the current frame,
if so, judging that the current action behavior of the player is 'kicking the ball';
judging whether opponent team players exist in a preset distance range from the player with the ball according to the identified and tracked player information and the football information,
if yes, judging that the player with the ball is currently in an attack and defense state;
for the ball kicking behavior of the players, whether the football far away from the ball kicker is got by other players is further judged,
if yes, judging that the football player is a team member or an opponent team member, if the football player obtains the ball, judging that the action behavior of the kickball player is 'pass ball', and if the football player obtains the ball, judging that the current action behavior of the opponent player is 'snap off';
if not, judging whether the football far away from the kickers approaches to the goal, and if so, judging that the kicking behavior of the kickers is 'shooting'.
The invention extracts the court features, the player features and the football features based on the color difference values of lawns, players, football, goals, referees and the like on the video frame image of the competition field, automatically identifies and tracks the players and the football on the subsequent frame based on the player features and the football features extracted from the first frame, reconstructs the three-dimensional animation scene of the football match according to the extracted court information, the identified and tracked player information and football information, and allows audiences to watch the match at any visual angle. In addition, the invention can automatically analyze the action of each player on the playing field according to the identified and tracked information of the players and the football information, and the analysis result can be used as a basis for quantitatively evaluating the performance of the players.
It should be understood that the above-described embodiments are merely preferred embodiments of the invention and the technical principles applied thereto. It will be understood by those skilled in the art that various modifications, equivalents, changes, and the like can be made to the present invention. However, such variations are within the scope of the invention as long as they do not depart from the spirit of the invention. In addition, certain terms used in the specification and claims of the present application are not limiting, but are used merely for convenience of description.
Claims (10)
1. A three-dimensional animation scene reconstruction system aiming at a football match video is characterized by comprising:
the football match video frame sequence input module is used for inputting a video frame sequence of a football match;
the court information extraction module is connected with the football match video frame sequence input module and is used for extracting court information representing court characteristics on each video frame;
the player identification tracking module is connected with the football match video frame sequence input module and is used for automatically identifying and tracking players on each video frame;
the football identification tracking module is connected with the football match video frame sequence input module and is used for automatically identifying and tracking the football motion track on each video frame;
the three-dimensional animation scene reconstruction module is respectively connected with the court information extraction module, the player identification tracking module and the football identification tracking module and is used for providing designers with three-dimensional animation scenes for reconstructing a football match according to the extracted court information, the identified and tracked player information and football information;
sportsman's action analysis module connects respectively sportsman discernment tracker with football discernment tracker for according to discernment, trail sportsman's information with football information analyzes sportsman's action, and output analysis result.
2. The system for reconstructing three-dimensional animation scene for video of football match according to claim 1, wherein the player identification tracking module comprises:
the player segmentation extraction unit is used for carrying out image segmentation on the video frame image so as to extract players on the video frame;
the team judgment unit is connected with the player division and extraction unit and used for judging the team to which the player belongs according to the extracted player characteristics to obtain a judgment result;
and the player motion tracking unit is connected with the player segmentation and extraction unit and used for realizing automatic tracking of the motion of each extracted player.
3. The system of claim 2, wherein the image segmentation algorithm based on color information performs image segmentation on the video frame image to extract players on the video frame.
4. The system for reconstructing three-dimensional animation scene for video of football match according to claim 3, wherein the image segmentation algorithm based on color information is expressed by the following formula:
in the above formula, b (x, y) represents the position of the suspected region of the player on the video frame;
x represents the coordinate of the suspected area position in the x-axis direction;
y represents the coordinate of the suspected area position in the y-axis direction;
g (x, y) represents a color value of a green component of the suspected location of the area;
r (x, y) represents a color value of a red component of the location of the suspected area;
b (x, y) represents a color value of a blue component of the location of the suspected area;
Trand TbTwo color thresholds;
b (x, y) ═ 1 indicates that the player at the position of the suspected area is true;
b (x, y) ═ 0 indicates that the player at the position of the area where the plausibility exists is false.
5. The system for reconstructing three-dimensional animation scene for video of football match according to claim 4, wherein the color threshold T isr＝15。
6. The system for reconstructing three-dimensional animation scene for video of football match according to claim 4, wherein the color threshold T isb＝60。
7. The system for reconstructing three-dimensional animation scene for football match video as claimed in claim 2, wherein the player action behavior analysis module specifically comprises:
a ball-carrying behavior judging unit for judging whether the distance between the football and the same player in a continuous frame is always kept in a preset ball distance range according to the identified and tracked player information and the football information,
if so, judging that the current action behavior of the player is 'with the ball';
a kicking behavior judging unit for judging whether the football is always in the control range of the football player in a continuous frame before the current frame and the football leaves the control range of the football player in the current frame according to the identified and tracked information of the football player and the football information,
if so, judging that the current action behavior of the player is 'kicking the ball';
the kicking behavior judging unit specifically comprises:
a pass behavior judging subunit for judging whether the football far away from the kickers is got by the team players,
if yes, judging the kicking behavior of the player as 'pass';
a snap action judgment subunit for judging whether the football far away from the kicker is got by the opponent team player,
if yes, judging the action behavior of the opposite player as 'snap';
a shooting behavior judging subunit, which is used for judging whether the football far away from the kickers approaches the goal or not,
if yes, judging the kicking behavior of the player as shooting;
the attack and defense state judging unit is connected with the ball-carrying behavior judging unit and used for judging whether opposite team players exist in a preset distance range from the ball-carrying players or not according to the identified and tracked player information and football information after judging that the current action behavior of the players is 'ball-carrying',
if yes, the player with the ball is judged to be in an attack and defense state at present.
8. A three-dimensional animation scene reconstruction method for a football match video is realized by applying the three-dimensional animation scene reconstruction system as claimed in any one of claims 1 to 7, and is characterized by comprising the following steps:
inputting a video frame sequence of a football game;
extracting court information on each video frame;
automatically identifying and tracking the players on each video frame;
identifying and tracking the football track on each video frame;
constructing a static three-dimensional scene of the football match according to the court information extracted from the first frame and the player position information and the football position information identified from the first frame;
mapping the tracked player motion information and the tracked football motion trail to the static three-dimensional scene, and reconstructing the three-dimensional animation scene of the football match.
9. A three-dimensional animation scene reconstruction method for a football match video is provided, wherein a player action behavior recognition method is realized by applying a three-dimensional animation scene reconstruction system as defined in any one of claims 1-7, and the method is characterized by comprising the following steps of:
inputting a video frame sequence of a football game;
extracting court information on each video frame;
automatically identifying and tracking the players on each video frame;
automatically identifying and tracking the football motion track on each video frame;
analyzing the action of the player according to the identified and tracked information of the player and the football information, and outputting an analysis result.
10. The method for reconstructing the three-dimensional animation scene aiming at the football match video and the identification method of the action behaviors of the players as claimed in claim 9 are characterized in that the specific method for analyzing the action behaviors of the players is as follows:
judging whether the distance between the football and the same player in a continuous frame is always kept in a preset ball distance range according to the identified and tracked player information and the football information,
if so, judging that the current action behavior of the player is 'with the ball';
according to the identified and tracked player information and football information, judging whether the football is always in the control range of the players in a continuous frame before the current frame and the football leaves the control range of the players in the current frame,
if so, judging that the current action behavior of the player is 'kicking the ball';
judging whether opponent team players exist in a preset distance range from the player with the ball according to the identified and tracked player information and the football information,
if yes, judging that the player with the ball is currently in an attack and defense state;
for the ball kicking behavior of the players, whether the football far away from the ball kicker is got by other players is further judged,
if yes, judging that the football player is a team member or an opponent team member, if the football player obtains the ball, judging that the action behavior of the kickball player is 'pass ball', and if the football player obtains the ball, judging that the current action behavior of the opponent player is 'snap off';
if not, judging whether the football far away from the kickers approaches to the goal, and if so, judging that the kicking behavior of the kickers is 'shooting'.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010804502.5A CN112200891A (en) | 2020-08-12 | 2020-08-12 | System and method for reconstructing video three-dimensional animation scene of football match |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010804502.5A CN112200891A (en) | 2020-08-12 | 2020-08-12 | System and method for reconstructing video three-dimensional animation scene of football match |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112200891A true CN112200891A (en) | 2021-01-08 |
Family
ID=74005571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010804502.5A Pending CN112200891A (en) | 2020-08-12 | 2020-08-12 | System and method for reconstructing video three-dimensional animation scene of football match |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112200891A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112883865A (en) * | 2021-02-09 | 2021-06-01 | 北京深蓝长盛科技有限公司 | Ball-bearing breakthrough event identification method and device, computer equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140177905A1 (en) * | 2012-12-20 | 2014-06-26 | United Video Properties, Inc. | Methods and systems for customizing a plenoptic media asset |
| CN104866853A (en) * | 2015-04-17 | 2015-08-26 | 广西科技大学 | Method for extracting behavior characteristics of multiple athletes in football match video |
| CN106474716A (en) * | 2016-06-24 | 2017-03-08 | 郑州动量科技有限公司 | A kind of footballer's pace work monitoring and evaluation method and its system |
| CN110826539A (en) * | 2019-12-09 | 2020-02-21 | 浙江大学 | Visual analytic system of football pass based on football match video |
-
2020
- 2020-08-12 CN CN202010804502.5A patent/CN112200891A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140177905A1 (en) * | 2012-12-20 | 2014-06-26 | United Video Properties, Inc. | Methods and systems for customizing a plenoptic media asset |
| CN104866853A (en) * | 2015-04-17 | 2015-08-26 | 广西科技大学 | Method for extracting behavior characteristics of multiple athletes in football match video |
| CN106474716A (en) * | 2016-06-24 | 2017-03-08 | 郑州动量科技有限公司 | A kind of footballer's pace work monitoring and evaluation method and its system |
| CN110826539A (en) * | 2019-12-09 | 2020-02-21 | 浙江大学 | Visual analytic system of football pass based on football match video |
Non-Patent Citations (1)
| Title |
|---|
| 杜威等: "足球比赛场景的三维再现和自动解说", 《计算机辅助设计与图形学学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112883865A (en) * | 2021-02-09 | 2021-06-01 | 北京深蓝长盛科技有限公司 | Ball-bearing breakthrough event identification method and device, computer equipment and storage medium |
| CN112883865B (en) * | 2021-02-09 | 2024-01-19 | 北京深蓝长盛科技有限公司 | Identification method and device for break-through event with ball, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102819749B (en) | A kind of football offside automatic discrimination system and method based on video analysis | |
| Thomas et al. | Computer vision for sports: Current applications and research topics | |
| US10824918B2 (en) | System and method for predictive sports analytics using body-pose information | |
| Pingali et al. | Real time tracking for enhanced tennis broadcasts | |
| US20180137363A1 (en) | System for the automated analisys of a sporting match | |
| CN111444890A (en) | Sports data analysis system and method based on machine learning | |
| JP6249706B2 (en) | Information processing apparatus, information processing method, and program | |
| US20220198684A1 (en) | Player Trajectory Generation via Multiple Camera Player Tracking | |
| MX2012000902A (en) | Play sequence visualization and analysis. | |
| Fu et al. | Screen-strategy analysis in broadcast basketball video using player tracking | |
| Chakraborty et al. | A trajectory-based ball detection and tracking system with applications to shot-type identification in volleyball videos | |
| Shishido et al. | A trajectory estimation method for badminton shuttlecock utilizing motion blur | |
| CN112200891A (en) | System and method for reconstructing video three-dimensional animation scene of football match | |
| US20230377336A1 (en) | Method of operating server providing sports video-based platform service | |
| CN115475373B (en) | Display method and device of motion data, storage medium and electronic device | |
| CN116758459A (en) | Real-time analysis system and method for badminton match | |
| Shishido et al. | Visual tracking method of a quick and anomalously moving badminton shuttlecock | |
| KR102612525B1 (en) | System, apparatus and method for master clock and composite image | |
| CN114463664A (en) | Novel ice hockey tracking method for ice hockey sports | |
| Chang et al. | Fencing tactics analysis in broadcast video: A point-by-point analytical system | |
| Siratanita et al. | A method of saliency-based football-offside detection using six cameras | |
| US9959632B2 (en) | Object extraction from video images system and method | |
| Gomez¹ et al. | Semi-automatic tracking of beach volleyball players | |
| Choroś | Improved video scene detection using player detection methods in temporally aggregated TV sports news | |
| Zhu et al. | Trajectory reconstruction system of moving target applied in volleyball match |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210108 |