JP2004072540A - Personal digest distribution system, personal digest preparing method and program for making computer execute the method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a personal digest distribution system capable of creating and distributing digest information by extracting information preferable to a user among scenes which are important and matched with the user's preference. <P>SOLUTION: A personal digest distribution system is provided with a preference information registering 109 for registering target preference indicating a target desired by the user to be distributed about images and registering situation preference indicating a target situation, a preference attribute determining module 117 for selecting a scene to be used a personal digest on the basis of the target preference and the situation preference registered by the user to whom the personal digest should be distributed, a significance calculation module 119 and a digest generation module 121. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a personal digest delivery system that creates and distributes digest information according to a user's preference, a personal digest creation method, and a program for causing a computer to execute the method.
[0002]
[Prior art]
In recent years, portable information terminals (portable terminals) such as PDA (Personal Digital Assistant) and mobile phones have been remarkably spread. In particular, in the third-generation mobile terminal, a service of receiving not only text and still images but also moving images has been started at a maximum of 384 Kbps. Furthermore, it is expected that a function of reading out by voice will be implemented in the future.
[0003]
By the way, there is a service for transmitting information about sports to a portable terminal. This service informs users of the progress of sports games such as baseball and soccer, and information on breaking results of the games. The user of the service can know the situation and the result of the game even when the user cannot watch the television or the like due to moving or the like.
[0004]
The service edits the highlights of the match, creates a digest of the match, and creates information to be sent to the mobile terminal. The digest information is composed of, for example, a character string and a scene corresponding to the character string such as “one back, team G3-1, team G, team G is reversed by a timely hit of team G player T”. In addition, since many mobile terminals are owned by individuals, a personal digest distribution system that automatically creates digest information reflecting preferences of each user has also been proposed.
[0005]
As a conventional example of a personal digest distribution system reflecting tastes, for example, a scene of high importance is selected from a video of a game based on information on the user's taste, and sorted as necessary in a time series for individuals. Some digest information is created. The importance of the video is determined by, for example, an event such as a home run or a reversal. Then, by selecting a scene that matches the user's preference (such as a favorite player) acquired in advance from the scenes extracted as having high importance, the user's preference is reflected in the importance, and is used for digest information. Scene to be selected.
[0006]
[Problems to be solved by the invention]
By the way, some users are disadvantaged to a player if the scene involves a favorite player, and some users want to see it as digest information regardless of their advantage. Some scenes do not want to see scenes such as strikeouts, errors, etc.). However, the above-described conventional digest information selects a scene regardless of disadvantage or advantage for a player, and thus cannot meet a user's request that he does not want to see a disadvantageous scene for a favorite player.
[0007]
The present invention has been made in view of the above points, and is an important scene for conveying the progress and results of a game, and among scenes that match the user's preference, more preferable information for the user It is an object of the present invention to provide a personal digest distribution system capable of extracting and generating digest information and distributing the digest information, a personal digest creation method, and a program for causing a computer to execute the method.
[0008]
[Means for Solving the Problems]
In order to achieve the above object, a personal digest distribution system according to claim 1 is a personal digest distribution system that creates and distributes a digest from a video composed of a plurality of continuous scenes. Target preference registering means for registering a target preference indicating a shooting target desired to be distributed; status preference registering means for registering a situation preference indicating a status of a shooting target desired by the user with respect to the shooting target; And scene selection means for selecting a scene used for a personal digest based on the target preference and the situation preference registered by a user to be delivered.
[0009]
According to the first aspect of the present invention, a target preference indicating a shooting target that a user desires to distribute with respect to a video is registered, and a situation indicating a status of a shooting target that the user desires to distribute with respect to the shooting target is registered. Preferences can be registered. Then, at the time of distributing the digest, it is possible to select a scene to be used for the personal digest based on the target preference and the situation preference registered by the user to be delivered. For this reason, it is possible to select a scene to be adopted in the digest by reflecting not only the user's preference regarding the shooting target but also the user's preference regarding the shooting target situation. Therefore, a digest scene more preferable for the user can be created and distributed.
[0010]
The personal digest distribution system according to the invention according to claim 2, wherein the situation preference is at least one of a situation in which the shooting target registered as the target preference is advantageous, an advantage for the shooting target, and a situation which is not disadvantageous. It is characterized by including.
[0011]
According to the second aspect of the present invention, the situation preference includes at least one of a situation in which the imaging target registered as the target preference is advantageous, an advantage for the imaging target, and a situation which is not disadvantageous. Therefore, it is possible to cope with both a user who wants to see only an advantageous situation for the photographing target and a user who wants to see the video relating to the photographing object regardless of whether it is advantageous or disadvantageous.
[0012]
The personal digest distribution system according to the invention according to claim 3, wherein the situation preference is such that when the video relays a sports game, the situation preference is an advantageous situation for the shooting target to attack, and the shooting target is It is characterized by including at least one of advantageous situations for defense.
[0013]
According to the third aspect of the present invention, the preference of the situation includes at least one of a situation that is advantageous for the target to be attacked and a situation that is advantageous for the target to be protected, so that the user has a better image for the user. You can create and distribute street digests.
[0014]
The personal digest distribution system according to the invention according to claim 4, further comprising video analyzing means for analyzing the meaning of the video, wherein the video analyzing means forms the video based on metadata for explaining the video. A parameter indicating the importance of the scene to be created, and an importance determining means for determining the importance of the scene based on the value of the parameter.
[0015]
According to the fourth aspect of the present invention, it is possible to create a parameter indicating the importance of a scene constituting a video based on the metadata, and determine the importance of the scene based on the value of the parameter. For this reason, it is possible to analyze the degree of importance of the video and determine a meaningful scene as a scene candidate to be used for the digest.
[0016]
The personal digest distribution system according to claim 5, wherein the importance determining unit determines the importance of a scene that matches the target preference and the situation preference, and determines the importance of each user. It is characterized by.
[0017]
According to the fifth aspect of the invention, it is possible to determine the importance for each of the users. For this reason, a digest can be created in a video scene more preferable for each user.
[0018]
7. The personal digest distribution system according to claim 6, wherein the importance determined by the importance determination unit according to the status indicating the position of the shooting target registered in the target preference in the scene and the situation preference. And a threshold for determining whether to select the scene or not.
[0019]
According to the invention described in claim 6, according to the status indicating the position in the scene to be photographed and the situation preference, the weight to be added to the importance determined by the importance determination means and whether to select the scene or not. In order to further provide a situation preference determining means for obtaining a threshold value for determining the situation preference, it is possible to change and adjust the scene selected according to the situation preference, and to create a digest considering the situation preference it can.
[0020]
The personal digest distribution system according to the invention of claim 7 obtains the weight according to the situation preference and each of the plurality of elements of the scene, and the weight corresponds to each of the plurality of elements of the scene. In addition, the apparatus further includes importance acquiring means for acquiring the final importance by adding each importance added to the importance and adding the weight.
[0021]
According to the invention described in claim 7, the weight is obtained according to each of the situation preference and the plurality of scene elements, and the weight is added to the importance corresponding to each of the plurality of scene elements. Since the weights are added to obtain the final importance, the final importance reflecting the importance of a plurality of elements of the scene can be obtained.
[0022]
9. The personal digest distribution system according to claim 8, wherein the scene selection unit ranks the scenes according to the final importance, and a scene having a final importance equal to or larger than the threshold among the scenes. And a digest creation unit that creates a digest by arranging the selected scenes in chronological order while selecting in the order of the highest final importance.
[0023]
According to the eighth aspect of the present invention, the scenes are ranked in accordance with the final importance, and the scenes having the importance equal to or higher than the threshold among the scenes are selected in descending order of importance, and the scenes are selected in time series. Since it is arranged, it is possible to create a digest using a scene with a final importance that is high considering multiple elements and that has a final importance equal to or greater than a threshold value that is determined in consideration of the user's situational preference. it can.
[0024]
10. The personal digest creation method according to claim 9, wherein: a target preference registration step of registering a target preference indicating a shooting target that a user desires to distribute with respect to a video composed of a plurality of continuous scenes; A situation preference registration step of registering a situation preference indicating the situation of the imaging target that the user desires to distribute with respect to the target, and at the time of digest distribution, based on the target preference and the situation preference registered by the user to be distributed Selecting a scene to be used for a personal digest.
[0025]
According to the ninth aspect of the present invention, a target preference indicating a shooting target that a user desires to distribute with respect to a video is registered, and a situation indicating a status of a shooting target that the user desires to distribute with respect to the shooting target is registered. Preferences can be registered. Then, at the time of distributing the digest, it is possible to select a scene to be used for the personal digest based on the target preference and the situation preference registered by the user to be delivered. For this reason, it is possible to select a scene to be adopted in the digest by reflecting not only the user's preference regarding the shooting target but also the user's preference regarding the shooting target situation. Therefore, a digest scene more preferable for the user can be created and distributed.
[0026]
According to a tenth aspect of the present invention, there is provided a program for causing a computer to execute the personal digest creating method of the ninth aspect.
[0027]
According to the tenth aspect, the personal digest creation method of the ninth aspect can be executed.
[0028]
BEST MODE FOR CARRYING OUT THE INVENTION
Exemplary embodiments of a personal digest distribution system, a personal digest creation method, and a program for causing a computer to execute the method according to the present invention will be described below in detail with reference to the accompanying drawings. In the present embodiment, a personal digest distribution system will be described with an example of creating and distributing a baseball digest.
[0029]
(Overview)
The personal digest distribution system according to the present embodiment is a system that creates a digest from video (video that relays a baseball game) composed of a plurality of continuous scenes. Then, in order to create a digest that is desirable for each user, the user registers in advance in the personal digest distribution system a video capturing target such as a team name or a player for whom the user wishes to distribute a personal digest. Furthermore, in the present embodiment, since a personal digest is created by editing only a scene desired by a player or team desired by an individual user, a situation in a video desired to be distributed to a shooting target is also registered.
[0030]
The above-described personal digest distribution system can create and distribute a personal digest according to each user's request, not only for a shooting target such as a team or a player, but also for a situation where the shooting target is placed.
[0031]
FIG. 1 is a diagram for explaining main parts of a personal digest distribution system according to one embodiment of the present invention. The personal digest distribution system 1 according to the present embodiment includes a data generation unit 103, a video content analysis unit 105, a digest generation unit 107, a preference information registration unit 109, a baseball database (DB) 111, and a user DB 112. Hereinafter, each configuration will be described.
[0032]
Preference information registration section
The preference information registration unit 109 according to the present embodiment registers a target preference (also referred to as a preference item) indicating a shooting target (a specific team or player in the present embodiment) that the user desires to distribute with respect to the video. And the situation preference indicating the situation of the shooting target (target) that the user wishes to distribute with respect to the shooting target (scored by team, player hit, or batter struck out, etc.) This is a situation preference registration unit to be registered. The preferences registered in the preference information registration unit 109 are stored in the user DB 112 together with information such as the address of the user.
[0033]
FIG. 2 is a diagram for explaining that the user registers a target preference and a situation preference in the preference information registration unit 109. The personal digest distribution system 1 transmits data for creating the screen shown in FIG. 2 to the mobile phone 115 as the mobile terminal device of the user U through the Internet I. Based on the transmitted data, a screen for registering the “favorite player” as the target of the target preference and the situation preference is displayed on the display screen 115a, which is the user interface (UIF) of the user U.
[0034]
The target is registered, for example, when the user specifies the following items in the preference information registration unit 109 using the UIF.
Favorite team
Favorite player
Note that a plurality of favorite teams and favorite players can be registered. When the registration of the target is completed, the characters “Register Team K No. 5N as a favorite player” are displayed on the display screen of the mobile phone 115.
[0035]
The situation preference includes at least one of a situation in which the target such as a team or a player registered as the target preference is advantageous for the target, a situation in which the target is advantageous or not disadvantageous. Also, as in the case of the present embodiment, when the video is a sport that relays a game, the situation preference is that the registered player or team has an advantage in attacking, and the registered player or team defends. At least one of the situations that is advantageous in doing so.
[0036]
The situation preference is registered by specifying one of the following items in the preference information registration unit 109 using the UIF, similarly to the target. The situation preference is also referred to as a preference attribute because it is registered as a preference attribute further than the target preference.
I want to see only N players' activities ... (1)
Focusing on good hits… (1.1)
Focusing on good defense… (1.2)
Focusing on both good points… (1.3)
I want to see the scenes of N players even if they are not active ... (2)
[0037]
Among the situation preferences described above, “I want to see only the active scenes of N players” shown in (1) indicates a situation (also referred to as a positive situation) in which the registered N players are advantageous for the shooting target. is there. Furthermore, "centering on good hits" shown in (1.1) is an advantageous situation for N players to attack, and "centering on good defenses" shown in (1.2) is , N players have an advantageous situation to defend. In addition, “I want to see the scene of N players even if they are not active” shown in (2) indicates a situation that is not advantageous or disadvantageous for N players.
[0038]
Note that, in the above example, an example in which an individual such as N players is an imaging target has been described. However, in the personal digest distribution system of the present embodiment, a team can be treated as an imaging target similarly to an individual. In this case, for example, only a scene that is positive for the team N, a scene that is positive when the team N attacks (hit, score, etc. of a player belonging to the team N), or a scene that is positive for defense (fine play of a player that belongs to the team N) , Strikeout, etc.) and register as situational preferences.
[0039]
Data generator
The data generation unit 103 inputs metadata in units of one play of a video of a game shot at the baseball field 2. The metadata is data for explaining a video. One play is, for example, a unit indicating the amount of data of a game centered on the last one ball of each at bat. Shows the progress of the match. In the present embodiment, it is assumed that the processing is performed by a site processing unit that is provided in the baseball stadium 2 and creates metadata according to the progress of the game and cuts out a video for one play corresponding to the metadata from the entire video. . Either the creation of the metadata or the clipping of the video may be performed first.
[0040]
The data generation unit 103 stores the input metadata in the baseball data DB 111. In addition, the data generation unit 103 inputs a video for one play corresponding to the metadata and edits the video into a video (video scene) having an average of about 15 seconds. The video scene is stored in the baseball data DB 111 in the same manner as the metadata.
[0041]
Video Content Analysis Department
The video content analysis unit 105 is configured to analyze the meaning of the video, creates a parameter indicating the importance of the video scene based on the metadata stored in the baseball DB 111, and calculates the importance based on the parameter value. . Then, the video scene determined to be important based on the calculated importance is stored in the baseball DB 111 as a digest candidate scene.
[0042]
Here, calculation of the importance of the video content analysis unit 105 will be described. The important parameters are set as a plurality of parameters according to each of the plurality of elements of the video scene. In the present embodiment, an attack parameter indicating an attack level, a pitcher parameter indicating a pitcher level, a defense parameter indicating a defense level, a situation level according to each of a plurality of elements (levels) such as an attack, a pitcher, a defense, and a situation of a scene Was set. The video content analysis unit 105 determines the content of the video in which the parameter value increases or decreases for each importance parameter, and stores it as data. FIG. 3 shows the importance parameter and the video content for which the parameter value is to be increased in association with each other.
[0043]
4 to 6 are diagrams for explaining how to calculate the parameter value of the importance parameter using the video contents shown in FIG. FIGS. 4A and 4B are diagrams for qualitatively explaining that the importance is determined based on the content of the video scene. FIGS. 5 and 6 show that the importance is calculated quantitatively. It is a figure for explaining.
[0044]
FIG. 4A shows the importance (parameter value) of a video scene including a scene in which the R player of the team K hits the center with a two-run home run. The importance of the video scene shown in FIG. 4A is, for example, a situation parameter such as “there is a runner in the scoring zone”, “the R player is a batter hitting a cleanup”, “R player home run, additional points”, “M” It is calculated based on attack parameters such as "player addition" and "team K, reversal". FIG. 4 (b) shows the importance of the video scene including the scene in which the K player of team L hits the third base liner, which was complemented by the N player of team K, and the team L rejected this round. Is shown. The importance of the video scene shown in FIG. 4B is divided into a situation parameter such as "K player is a batter hitting a cleanup", a defensive parameter such as "good catching ball", and a pitcher parameter such as "batter retreat, three retreat". It is calculated based on:
[0045]
Next, a method of calculating the importance parameter value will be described with reference to FIGS. In FIG. 5, the importance of the attack level of the video scene included in one play is from the start of a turn at bat of a certain player, the advance (base) by hit, and the result of this hit becoming a timely hit and the point being added. An example of calculating the degree will be described. The importance is calculated using the attack parameters defined in FIG. The attack parameter represents an aggressive feature of the video scene, and is attached to an event that is aggressively important as it has a larger value.
[0046]
In the video scene shown in FIG. 5, a hit event is added with a 4 × attack coefficient as an attack parameter as shown in FIG. In addition, when this hit is timely, 2 × attack coefficient is added, and when the hit is added, 2 × attack parameter is added. The importance of the video scene shown in FIG. 5 is obtained by adding all attack parameters (4 × attack coefficient + 2 × attack coefficient + 2 × attack coefficient).
[0047]
The attack coefficient is defined by the following equation.
Attack coefficient:
Var attack coefficient = [/. * / [Attack coefficient = 1 + inning number × cleanup coefficient]] The cleanup coefficient is set to, for example, 0.7 when the batter is a player who hits the cleanup, and is set to 0 for other batters. . According to the above-mentioned attack coefficient, the attack parameter becomes larger as the inning of the game progresses, and the batter of the batter who hits the cleanup has a larger attack parameter than the batter of other batters.
[0048]
As described above, the video content analysis unit 105 calculates an attack coefficient and obtains an attack parameter of each event. Then, the attack parameters of all the events included in the video scene are added to calculate the importance of the attack level of the video scene. Similarly, the importance of pitcher parameters, defensive parameters, and situation parameters is calculated, and the final importance and characteristics of each video scene can be determined by calculating the values of the four importance parameters. . Each video scene whose importance has been calculated is stored in the baseball DB 111.
[0049]
Digest generation unit
At the time of distribution of the personal digest, the digest generation unit 107 selects a video scene to be used for the personal digest based on the target preference and the situation preference registered by the distribution destination user. The digest generation unit 107 includes a preference attribute determination module 117, an importance calculation module 119, and a digest generation module 121.
[0050]
The preference attribute determination module 117 determines the status in the video scene of the target registered as the target preference (the position of the target registered as the target preference; in this embodiment, the attacker, the defensive player, or the pitcher). Or the fielder), a video scene that is advantageous to the target, and a weight that is added to the importance of the video scene according to the situation preference such as whether or not it is advantageous in an attack, and the weight is added. In this configuration, a threshold value for determining whether or not to select a video scene in comparison with the importance level is obtained.
[0051]
Also, the importance calculation module 119 obtains a weight according to each of a plurality of elements of the video scene, such as the status, the preference of the situation, the attack level, the pitcher level, the defensive level, and the status level, and analyzes the weight by the video content analysis. This is an importance obtaining unit that adds the importance corresponding to each of the parameter levels calculated by the unit 105 and adds the weighted importance to obtain the final importance.
[0052]
Further, the digest generation module 121 ranks the video scenes stored in the baseball DB 111 in accordance with the final importance, and determines a scene having a final importance equal to or more than a threshold among the ranked video scenes to the final importance. The digest is created by selecting in descending order and arranging them in chronological order.
[0053]
More specifically, the preference attribute determination module 117 determines the status of the target (for example, N players) based on the metadata for each video scene. Then, the status and the preference are input, and the weight (importance coefficient) added to the importance of each video scene is calculated. Further, a threshold value for selecting whether to use the video scene for the digest is calculated. The analyze_pref_attr, which is a function for the preference attribute determination module 117 to calculate the importance coefficient and the threshold, is expressed as follows.
analyze_pref_attr (metadata, pref_item, pref_attr) = (w1,... wn, t)
metadata: metadata of each video scene
pref_item: target preference
pref_attr: Situation preference
[0054]
Note that the analyze_pref_attr is activated for each video scene. Further, the above-described importance coefficient and threshold value are calculated for each of the aggressive level, the pitcher level, the defensive level, and the situation level. For this reason, the preference attribute determination module 117 assigns a weight corresponding to each level to four importance factors (denoted as α, β, γ, and δ in the present embodiment) and one threshold (the present embodiment). Then, write θ).
[0055]
The preference attribute determination module 117 of the present embodiment manages the above processing by managing the table shown in FIG. 7 and according to each situation preference (preference attribute) for each status of the target registered as the target preference. The threshold value and the importance coefficient are stored. Then, the threshold value and the importance coefficient are returned to the importance calculation module 119 in accordance with the arguments determined from the metadata, such as the status, the object preference, and the situation preference.
[0056]
In FIG. 7, importance factors α, β, γ, δ, and threshold value θ are further determined for each situation preference (preference attribute), and α, β, γ, δ are further determined as α1 to α4, β1 to β4. β4, γ1 to γ4, and θ1 to θ4 are set separately. When the situation preference is, for example, offensive positive, the importance level coefficient α1 of the attack level is set to a value larger than other levels. Further, the importance coefficient β1 of the defense level and the importance coefficient γ1 of the pitcher level may be set to a value smaller than other levels, or set to a negative value.
[0057]
The importance calculation module 119 multiplies the importance of the video scene stored in the baseball DB 111 by the importance coefficient set in the preference attribute determination module 117 to calculate the final importance (final importance). An equation for calculating the final importance is shown below.
[0058]
Final importance = Σ (α1 × attack parameter value + β1 × defensive parameter value + γ1 × pitcher parameter value + δ1 × situation parameter value) ... Equation (1)
Therefore, in the importance calculation module 119, the final importance of a video scene including an event having a large parameter value relating to an attack level or a video scene including a large number of events relating to a target attack is set to a large value.
[0059]
The digest generation module 121 ranks the video scenes in accordance with the final importance, and determines a scene having a final importance equal to or more than the threshold set by the preference attribute determination module 117 among the ranked video scenes as a final importance. Select in descending order. For this reason, the importance value of the video scene that is aggressively advantageous to the subject increases, and the digest generation module 121 preferentially selects a video scene that matches the user's preference. Further, the digest generation module 121 generates a digest by arranging the selected video scenes in chronological order.
[0060]
Here, in order to explain the effect of the digest distribution system of the present embodiment, a method of calculating importance in the related art will be briefly described. In the conventional digest distribution system, the importance of each video scene is obtained using the following equation (2).
Final importance = Σ (attack parameter value + defensive parameter value + pitcher parameter value + situation parameter value) ... Equation (2)
[0061]
According to equation (2), the final importance similarly increases when any of the parameter values, such as the attack parameter value and the defensive parameter value, is large. For this reason, in the conventional personal digest distribution system, when the user's target preference is, for example, team G, the team G is targeted regardless of the status of the team G such as the attacking side and the defending side, and the defense and pitcher's Higher video scenes are also employed in the digest.
[0062]
Therefore, according to the digest distribution system of the present embodiment described above, according to the situation preference of each user such as aggressive positive (advantageous when the target attacks) and other (irrespective of the advantage of the target). Weighting becomes possible. Then, the weighted importance is compared with a threshold value corresponding to the preference of the situation, a video scene is selected, and a digest can be created. For this reason, it is possible to create a digest that reflects not only the preference of the target such as the user's favorite team or player, but also the preference regarding the situation in which the target is advantageous in the attack or whether the target is advantageous or disadvantageous. Therefore, it is possible to create and distribute a digest suitable for every user.
[0063]
FIG. 8 is a flowchart for explaining a digest creation method performed by the above-described digest distribution system. The illustrated flowchart starts when the user accesses the digest distribution system in hope of the digest distribution. When there is access, the digest generation unit 107 reads the target preference (preference item) and the situation preference (preference attribute) of the accessed user stored in the user DB 112 in the preference attribute determination module 117 (step S801). ).
[0064]
Next, the preference attribute determination module 117 determines whether or not the user DB 112 has information on the user's preference (step S802). As a result, if the number of information items is n (step S802: n> 0), the number of processes i is initialized, and video scenes related to preference items are searched for all n pieces of information (steps S803 and S804). If it is determined in step S802 that the number of information items n is 0 (step S802: n = 0), a signal for instructing that information relating to preference is not registered is output to the outside (step S813). ).
[0065]
Next, as a result of the search in step S804, the preference attribute determination module 117 determines whether or not a video scene related to the registered preference information exists in the baseball DB 111 (step S805). As a result, when it is determined that there is no video scene in the baseball DB 111 (step S805: m = 0), a signal for instructing that there is no video relating to the target is output to the outside (step S814). On the other hand, if it is determined that there are m images relating to the object (step S805: m> 0), the number of times of processing i is reset, and the status of the object is determined until i reaches m (steps S806 and 807).
[0066]
Next, the preference attribute determination module 117 obtains an importance coefficient of the importance parameter based on the determined status (step S808). Then, the importance factor is multiplied by the importance parameter and added to calculate a final value of importance (step S809).
[0067]
Further, when it is determined in step S803 that i has reached n (step S803: i = n), the video scenes for which the final importance has been calculated are ranked in the order of the final importance (step S810). Then, among the ranked video scenes, the video scenes having the final importance not less than the threshold value are selected and arranged in time series (sort) (step S811). The sorted video scenes are distributed as digests to users who have requested distribution (step S812).
[0068]
The program that causes the computer to execute the personal digest distribution method of the personal digest distribution system of the present embodiment is a file in an installable or executable format, such as a CD-ROM, a floppy (R) disk (FD), It is provided by being recorded on a computer-readable recording medium such as a DVD.
[0069]
【The invention's effect】
As described above, the invention described in claim 1 selects a scene to be adopted in the digest by reflecting not only the preference of the user regarding the shooting target for each user but also the preference of the user regarding the situation of the shooting target. This provides an effect of providing a personal digest distribution system that can perform the above.
[0070]
The invention described in claim 2 is applicable to both a user who wants to see only a situation that is advantageous for the photographing target, and a user who wants to see the video relating to the photographing object regardless of disadvantages. An advantage is provided in that a personal digest distribution system that can select a scene adopted in a digest by reflecting even the user's preference regarding the target situation can be provided.
[0071]
The invention described in claim 3 has an effect that it is possible to provide a personal digest distribution system capable of creating and distributing a digest more exactly as an image for the user.
[0072]
According to the fourth aspect of the present invention, since the significance of a video is analyzed and a meaningful scene can be used as a candidate for a scene used for a digest, a process of selecting a video scene used for a personal digest can be simplified, An advantage is provided in that a personal digest distribution system capable of distributing a personal digest in a short time by user access can be provided.
[0073]
The invention described in claim 5 has an effect that it is possible to provide a personal digest distribution system that can create a digest in a more preferable video scene for each user.
[0074]
According to the sixth aspect of the present invention, the scene to be selected can be changed and adjusted according to the situation preference, so that the scene to be adopted in the digest can be selected by reflecting even the user preference regarding the situation of the shooting target. This has the effect of providing a personal digest delivery system that can be used.
[0075]
According to the seventh aspect of the present invention, since the final importance reflecting the importance of a plurality of elements of the scene can be acquired, it is adopted in the digest by reflecting more detailed user preference regarding the situation of the shooting target. This provides an effect of providing a personal digest distribution system capable of selecting a scene to be played.
[0076]
According to an eighth aspect of the present invention, a digest using a scene having a high final importance in consideration of a plurality of elements and having a final importance equal to or higher than a threshold obtained in consideration of a user's situational preference is used. Therefore, there is an effect that it is possible to provide a personal digest distribution system capable of selecting a scene adopted in a digest by reflecting a more detailed user preference regarding a situation of an imaging target.
[0077]
According to the ninth aspect of the present invention, a personal digest creation that can select a scene to be adopted as a digest by reflecting not only the preference of the photographing target for each user but also the preference of the user regarding the situation of the photographing target. There is an effect that a method can be provided.
[0078]
The invention according to claim 10 has an effect that the method for creating a personal digest according to claim 9 can be executed.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a main part of a personal digest distribution system according to an embodiment of the present invention.
FIG. 2 is a diagram for explaining that a user registers a target preference and a situation preference in a preference registration unit.
FIG. 3 is a diagram showing an association between an importance parameter and a video content for which the parameter value is increased;
FIG. 4 is a diagram for explaining how to calculate parameter values of importance parameters using the video contents shown in FIG. 3;
FIG. 5 is another diagram for explaining calculation of a parameter value of an importance parameter using the video content shown in FIG. 3;
FIG. 6 is another diagram for explaining calculation of a parameter value of an importance parameter using the video content shown in FIG. 3;
FIG. 7 is a diagram illustrating a table managed by a preference attribute determination module.
FIG. 8 is a flowchart illustrating a personal digest creation method executed in the personal digest distribution system according to one embodiment of the present invention.
[Explanation of symbols]
1 Personal digest delivery system
2 Baseball field
103 Data generation unit
105 Video Content Analysis Unit
107 digest generation unit
109 Preference Register
111 Baseball DB
112 User DB
115 mobile phone
115a Display screen
117 Preference attribute determination module
119 Importance calculation module
121 digest generation module

Claims (10)

A personal digest distribution system that creates a digest from a video composed of a plurality of continuous scenes and distributes the digest,
A target preference registration unit for registering a target preference indicating a shooting target that a user desires to distribute with respect to the video,
Situation preference registration means for registering a situation preference indicating the situation of the imaging target that the user wishes to distribute with respect to the imaging target,
When delivering a digest, scene selection means for selecting a scene used for a personal digest based on the target preference and the situation preference registered by a user to be delivered,
A personal digest distribution system comprising: 2. The personal digest according to claim 1, wherein the situation preference includes at least one of a situation in which the imaging target registered as the target preference is advantageous, an advantage for the imaging target, and a situation that is not disadvantageous. 3. Delivery system. When the video is a relay of a sports game, the situation preference includes at least one of a situation that is advantageous for the shooting target to attack and a situation that is advantageous for the shooting target to defend. 3. The personal digest distribution system according to claim 2, wherein: The image analysis device further includes image analysis means for analyzing the meaning of the image, wherein the image analysis means creates a parameter indicating a degree of importance of a scene constituting the image based on metadata for explaining the image, and The personal digest distribution system according to any one of claims 1 to 3, further comprising an importance determining unit configured to determine importance of the scene based on the value. The personal digest distribution system according to claim 4, wherein the importance determining unit determines the importance of a scene that matches the target preference and the situation preference, and determines the importance of each user. According to the status indicating the position of the shooting target registered in the target preference in the scene and the situation preference, the weight added to the importance determined by the importance determination unit and whether to select the scene or not. The personal digest distribution system according to any one of claims 1 to 5, further comprising a situation preference judging means for acquiring a threshold for determination. Acquiring the weight according to the situation preference and each of the plurality of elements of the scene, and adding the weight to the importance corresponding to each of the plurality of elements of the scene; 7. The personal digest distribution system according to claim 6, further comprising an importance obtaining unit configured to obtain the final importance by adding the degrees. The scene selecting means ranks the scenes according to the final importance, and selects scenes having final importance equal to or larger than the threshold value from the scenes in descending order of final importance. The personal digest distribution system according to claim 7, further comprising a digest creating unit that creates a digest by arranging the digests in a time-series manner. A target preference registration step of registering a target preference indicating a shooting target that the user desires to distribute with respect to a video composed of a plurality of continuous scenes,
A situation preference registration step of registering a situation preference indicating the situation of the imaging target that the user desires to distribute with respect to the imaging target,
When delivering a digest, a scene selection step of selecting a scene used for a personal digest based on the target preference and the situation preference registered by a user to be delivered,
A personal digest creation method, comprising: 10. A program for causing a computer to execute the personal digest creation method according to claim 9.

Images