CN110955786A - Dance action data generation method and device - Google Patents
Dance action data generation method and device Download PDFInfo
- Publication number
- CN110955786A CN110955786A CN201911206883.0A CN201911206883A CN110955786A CN 110955786 A CN110955786 A CN 110955786A CN 201911206883 A CN201911206883 A CN 201911206883A CN 110955786 A CN110955786 A CN 110955786A
- Authority
- CN
- China
- Prior art keywords
- dance
- phrase
- phrases
- music
- target music
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
- G06F16/432—Query formulation
- G06F16/433—Query formulation using audio data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/48—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/489—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using time information
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Library & Information Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
The application relates to the technical field of dance arrangement, in particular to a dance action data generation method and device. The dance action data matched with the target music can be generated by dividing the acquired target music of the dance action data to be matched into a plurality of phrases, determining dance sentences matched with the phrases according to the characteristic information of the phrases, the time sequence information of the phrases in the target music and pre-stored dance sentence characteristic information corresponding to the phrases respectively, wherein the dance sentences matched with any two phrases in the preset time interval range of the target music are different, and splicing the dance sentences matched with the phrases respectively according to the time sequence information of the phrases. By adopting the scheme, the diverse dance action data matched with the target music can be quickly generated without model training, the generation efficiency is high, and the matching flexibility is high.
Description
Technical Field
The application relates to the technical field of dance arrangement, in particular to a dance action data generation method and device.
Background
Dances matched with given music are generated according to the given music and widely applied to dance games and animation production. In the traditional dance game or animation production process, the dance game or animation production process is generally divided into three steps, wherein in the first step, a professional dance engineer performs dance motion design and arrangement according to given music; secondly, dance movements are presented by professional dance actors, and meanwhile dance movements are captured by means of a movement capturing tool; and thirdly, performing post-processing on each frame of captured images of the dance movements by the animator to obtain the dance matched with the given music. By adopting the dance method, a short period of dance takes a long time to finish the processing, and the choreography of the dance is closely related to music, so that the three steps are required to be repeated for any change of music and actions, and the labor cost and the time cost are greatly increased.
The existing machine learning dance algorithm usually presets similar music corresponding to the same dance, so that dance is matched by calculating the similarity degree of given music and music in a data set, and by adopting the dance algorithm, on one hand, a large amount of training data is needed; on the other hand, in the actual dance process, the same dance can correspond to completely different music, and the dance matching method for the given music according to the similarity degree of the music enables the matched dance to be relatively single and low in flexibility.
Disclosure of Invention
In view of this, embodiments of the present application provide at least a method and an apparatus for generating dance motion data, which can quickly generate diverse dance motion data matched with target music, and have high generation efficiency and high matching flexibility.
Mainly comprises the following aspects:
in a first aspect, an embodiment of the present application provides a dance action data generation method, where the dance action data generation method includes:
obtaining target music of dance action data to be matched, and dividing the target music into a plurality of phrases;
determining a dance sentence matched with each phrase according to the characteristic information of each phrase in the phrases, the time sequence information of each phrase in the target music and pre-stored dance sentence characteristic information corresponding to the phrases; the dance sentences matched with any two phrases in the preset time interval range of the target music are different;
and splicing the dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases to generate dance motion data matched with the target music.
In one possible implementation manner, the obtaining of the target music of the dance motion data to be matched includes:
and acquiring the target music to be matched with the dance action data in a game scene or animation scene of the terminal.
In one possible embodiment, the target music is segmented into a plurality of phrases according to the following steps:
determining a plurality of repetitive music structures in the target music based on the characteristic information of the rhythm point in the target music; each repetitive music structure is determined by the characteristic information of a plurality of continuous rhythm points;
according to the multiple repetitive music structures, the target music is cut into multiple sections of music, and the dividing position of a dividing point between any two adjacent sections of music in the multiple sections of music is determined;
adjusting the dividing positions of the dividing points according to the beat information of the target music, and adding at least one dividing compensation point;
and cutting the target music into the plurality of phrases according to the division point after the division position is adjusted and the at least one division compensation point.
In a possible implementation manner, the determining a plurality of repetitive music structures in the target music based on the feature information of the rhythm point in the target music includes:
calculating a feature vector of a rhythm point in the target music; each element in the feature vector represents the similarity between the rhythm point and each rhythm point in the target music;
forming a feature matrix by the feature vectors of all the rhythm points;
and determining a plurality of repetitive music structures in the target music based on the feature matrix.
In one possible embodiment, the dividing the target music into a plurality of pieces of music according to the plurality of repetitive music structures includes:
clustering the plurality of repetitive music structures to obtain a clustering result;
and according to the clustering result, segmenting the target music into the plurality of sections of music.
In one possible embodiment, the plurality of dance sentences are pre-stored in a dance database; the generating method further comprises generating the dance database according to the following steps:
obtaining multi-section dances;
aiming at each dance in the multi-section dances, dividing each dance into a plurality of dancing sentences according to the characteristic information of each dance action in each dance;
and generating the dance database based on the segmented dance sentences and the dance action characteristic information corresponding to each dance sentence.
In a possible implementation manner, the determining a dance sentence matched with each phrase according to feature information of each phrase in the phrases, timing information of each phrase in the target music, and pre-stored dance sentence feature information respectively corresponding to the phrases comprises:
aiming at each phrase in the phrases, calculating the matching degree of the phrase and each dance phrase in the phrases according to the characteristic information of the phrase, the time sequence information of the phrase and the pre-stored dance phrase characteristic information corresponding to the phrases;
and determining the dance sentences matched with the phrases based on the matching degree of each phrase in the phrases and the dance sentences.
In a possible embodiment, the determining a dance sentence matched with each phrase based on a matching degree of each phrase of the phrases and the phrases, includes:
obtaining a plurality of first candidate matching dancing sentences corresponding to first phrases in the target music;
determining a plurality of second candidate matching dances corresponding to a second phrase according to the matching degrees respectively corresponding to the second phrase and the plurality of dances after the first phrase and the incidence relation between each first candidate matching dance in the plurality of first candidate matching dances and the plurality of dances; the first phrase and the second phrase are any two adjacent phrases in the target music;
and selecting the first candidate matching dancing sentence with the least second candidate matching dancing sentence as the dancing sentence matched with the first phrase from the plurality of first candidate matching dancing sentences.
In a second aspect, an embodiment of the present application further provides a dance motion data generation apparatus, where the dance motion data generation apparatus includes:
the dance action data matching module is used for matching dance action data with dance action data to obtain target music;
the determining module is used for determining a dance sentence matched with each phrase according to the characteristic information of each phrase in the phrases, the time sequence information of each phrase in the target music and pre-stored dance sentence characteristic information corresponding to each phrase; the dance sentences matched with any two phrases in the preset time interval range of the target music are different;
and the first generating module is used for splicing the dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases to generate dance action data matched with the target music.
In a possible implementation manner, the obtaining module is specifically configured to obtain the target music according to the following steps:
and acquiring the target music to be matched with the dance action data in a game scene or animation scene of the terminal.
In one possible implementation, the obtaining module includes:
the first determining unit is used for determining a plurality of repetitive music structures in the target music based on the characteristic information of the rhythm point in the target music; each repetitive music structure is determined by the characteristic information of a plurality of continuous rhythm points;
the first dividing unit is used for dividing the target music into a plurality of sections of music according to the plurality of repetitive music structures and determining the dividing position of a dividing point between any two adjacent sections of music in the plurality of sections of music;
the adjusting unit is used for adjusting the dividing positions of the dividing points according to the beat information of the target music and adding at least one dividing compensation point;
and the second segmentation unit is used for segmenting the target music into the multiple phrases according to the segmentation point after the segmentation position is adjusted and the at least one segmentation compensation point.
In a possible implementation, the first determining unit is configured to determine the repetitive music structure according to the following steps:
calculating a feature vector of a rhythm point in the target music; each element in the feature vector represents the similarity between the rhythm point and each rhythm point in the target music;
forming a feature matrix by the feature vectors of all the rhythm points;
and determining a plurality of repetitive music structures in the target music based on the feature matrix.
In a possible implementation, the first dividing unit is configured to divide the target music into a plurality of pieces of music according to the following steps:
clustering the plurality of repetitive music structures to obtain a clustering result;
and according to the clustering result, segmenting the target music into the plurality of sections of music.
In one possible embodiment, the plurality of dance sentences are pre-stored in a dance database; the generating device further comprises a second generating module; the second generation module is used for generating the dance database according to the following steps:
obtaining multi-section dances;
aiming at each dance in the multi-section dances, dividing each dance into a plurality of dancing sentences according to the characteristic information of each dance action in each dance;
and generating the dance database based on the segmented dance sentences and the dance action characteristic information corresponding to each dance sentence.
In one possible embodiment, the determining module includes:
the calculating unit is used for calculating the matching degree of each phrase and each dance sentence in the dance sentences according to the feature information of the phrase, the time sequence information of the phrase and pre-stored dance sentence feature information corresponding to the dance sentences;
and the second determining unit is used for determining the dancing sentences matched with the phrases on the basis of the matching degrees of the phrases and the dancing sentences.
In a possible embodiment, the second determining unit is configured to determine a dance sentence matching each phrase according to the following steps:
obtaining a plurality of first candidate matching dancing sentences corresponding to first phrases in the target music;
determining a plurality of second candidate matching dances corresponding to a second phrase according to the matching degrees respectively corresponding to the second phrase and the plurality of dances after the first phrase and the incidence relation between each first candidate matching dance in the plurality of first candidate matching dances and the plurality of dances; the first phrase and the second phrase are any two adjacent phrases in the target music;
and selecting the first candidate matching dancing sentence with the least second candidate matching dancing sentence as the dancing sentence matched with the first phrase from the plurality of first candidate matching dancing sentences.
In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, the processor and the memory communicate with each other through the bus when the electronic device runs, and the machine-readable instructions are executed by the processor to perform the steps of the dance motion data generation method according to the first aspect or any one of the possible embodiments of the first aspect.
In a fourth aspect, this application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the dance motion data generation method described in the first aspect or any one of the possible implementation manners of the first aspect are performed.
In the embodiment of the application, the obtained target music of the dance action data to be matched is divided into a plurality of phrases, the phrases matched with the phrases can be determined according to the characteristic information of the phrases, the time sequence information of the phrases in the target music and the pre-stored dance phrase characteristic information corresponding to the phrases, wherein the phrases matched with any two phrases in the preset time interval range of the target music are different, and the phrases matched with the phrases are spliced according to the time sequence information of the phrases, so that the dance action data matched with the target music can be generated. By adopting the scheme, the diverse dance action data matched with the target music can be quickly generated without model training, the generation efficiency is high, and the matching flexibility is high.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
FIG. 1 is a flow chart illustrating a method for generating dance motion data according to an embodiment of the present application;
FIG. 2 is a flow chart of another dance motion data generation method provided in the embodiment of the present application;
FIG. 3 is a functional block diagram of an apparatus for generating dance motion data according to an embodiment of the present application;
FIG. 4 shows a functional block diagram of the acquisition module of FIG. 3;
FIG. 5 is a second functional block diagram of a dance motion data generating apparatus according to an embodiment of the present application;
FIG. 6 is a functional block diagram of the determination module of FIG. 3;
fig. 7 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.
Detailed Description
To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Additionally, it should be understood that the schematic drawings are not necessarily drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and that steps without logical context may be performed in reverse order or concurrently. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.
In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
To enable those skilled in the art to utilize the present disclosure, the following embodiments are presented in conjunction with a particular application scenario, "game scenario or animation scenario," and it will be apparent to those skilled in the art that the general principles defined herein may be applied to other embodiments and application scenarios without departing from the spirit and scope of the present disclosure.
The method, the apparatus, the electronic device, or the computer-readable storage medium described in the embodiments of the present application may be applied to any scenario in which dance motion data needs to be generated, and the embodiments of the present application do not limit a specific application scenario, and any scheme using the method and the apparatus for generating dance motion data provided in the embodiments of the present application is within the scope of protection of the present application.
It is noted that prior to the present application, in a conventional dance game or animation process, there were generally three steps, the first step, a professional dance engineer designing and arranging the dance movements according to given music; secondly, dance movements are presented by professional dance actors, and meanwhile dance movements are captured by means of a movement capturing tool; and thirdly, performing post-processing on each frame of captured images of the dance movements by the animator to obtain the dance matched with the given music. By adopting the dance method, a short period of dance takes a long time to finish the processing, and the choreography of the dance is closely related to music, so that the three steps are required to be repeated for any change of music and actions, and the labor cost and the time cost are greatly increased.
Here, in the existing machine learning dance algorithm, similar music is usually preset to correspond to the same dance, so that the dance is matched by calculating the similarity degree of the given music and the music in the data set, and by adopting the dance algorithm, a large amount of training data is needed on one hand; on the other hand, in the actual dance process, the same dance can correspond to completely different music, and the dance matching method for the given music according to the similarity degree of the music enables the matched dance to be relatively single and low in flexibility.
In view of the above problems, in the embodiment of the application, the target music of the dance motion data to be matched is divided into a plurality of phrases, and the dance sentences matched with the phrases can be determined according to the characteristic information of each phrase, the time sequence information of each phrase in the target music, and pre-stored dance sentence characteristic information corresponding to the plurality of dance sentences, wherein the dance sentences matched with any two phrases in the preset time interval range of the target music are different, and the dance sentences matched with the phrases are spliced according to the time sequence information of each phrase, so that dance motion data matched with the target music can be generated. By adopting the scheme, the diverse dance action data matched with the target music can be quickly generated without model training, the generation efficiency is high, and the matching flexibility is high.
It should be noted that a phrase is a characteristic basic structural unit constituting a piece of music, and has a music melody fragment with relatively complete music image; the dance sentence is the minimum unit capable of independently and relatively completely expressing the intention of the author, and the dance action sequence capable of relatively and completely expressing certain emotion, aesthetic feeling and connotation can be realized.
For the convenience of understanding of the present application, the technical solutions provided in the present application will be described in detail below with reference to specific embodiments.
Fig. 1 is a flowchart of a dance action data generation method according to an embodiment of the present application. As shown in fig. 1, a method for generating dance motion data provided in an embodiment of the present application includes the following steps:
s101: and acquiring target music of dance action data to be matched, and dividing the target music into a plurality of phrases.
In a specific implementation, target music needing dance motion data matching is acquired, and the acquired target music is processed to match dance motion data with the target music, wherein the processing means dividing the target music into a plurality of phrases.
Here, in a game scene or animation scene of the terminal, target music to be matched with dance motion data may be acquired.
The scheme can be applied to dance-type game scenes and animation production scenes, and in the scenes, after the target music is acquired, diverse dance action data matched with the target music can be quickly generated.
S102: determining a dance sentence matched with each phrase according to the characteristic information of each phrase in the phrases, the time sequence information of each phrase in the target music and pre-stored dance sentence characteristic information corresponding to the phrases; and the dance sentences matched with any two phrases in the preset time interval range of the target music are different.
In a specific implementation, when dance motion data is matched for target music, dance sentences can be matched for each of multiple phrases in the target music respectively, specifically, for each phrase, the dance sentence matched with the phrase is determined according to the characteristic information of the phrase, the time sequence information of the phrase in the target music and the pre-stored characteristic information of each phrase in the multiple dance sentences, and the dance sentences matched with any two phrases in the preset time interval range of the target music are different.
Here, the feature information of each phrase includes tone information, pitch information, rhythm information, and the like of the phrase; the time sequence information of each phrase in the target music comprises the position of the phrase in the target music, the starting time and the ending time of the phrase in the target music, the duration of the phrase and the like, wherein the position is the first phrase in the target music; the dance sentence characteristic information of each dance sentence comprises characteristic information of each dance action included in the dance sentence, and the dance sentence characteristic information comprises dance types, action speeds, time sequence information of the dance sentences in the original dance and the like; the preset time interval range of the target music can be understood as that the dancing sentences matched with any two phrases are required to be different in the total time range corresponding to a plurality of continuous phrases in the target music, for example, the dancing sentences matched with the phrases of the continuous preset number cannot be repeated, so that the repetition rate of the dancing sentences matched with the phrases can be reduced, and the matching effect of dancing is improved.
It should be noted that, the existing machine learning dancing algorithm matches dances by calculating the similarity degree of given music and music in a data set, needs a large amount of training data, and makes the dances matched to be single, and the scheme proposes a new scheme for matching dancing sentences by phrases, only needs target music of dance motion data to be matched, does not need a large amount of training music data, matches dancing sentences for each phrase, and can match dancing sentences for each phrase only according to the characteristic information of the phrase, the time sequence information of the phrase at the target music, and pre-stored characteristic information of the corresponding dancing sentences, so as to quickly generate diverse dance motion data matched with the target music, the scheme has small calculation amount, high calculation speed, high generation efficiency, and high matching flexibility, can ensure the matching of a certain artistic level.
S103: and splicing the dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases to generate dance motion data matched with the target music.
In specific implementation, after the dance sentences are matched for each phrase in the target music, the dance sentences matched with the phrases can be spliced in sequence from the first phrase to the last phrase of the target phrase according to the time sequence information of each phrase in the target music to generate dance motion data matched with the target music, and then dance animation images can be generated according to the dance motion data and the target music. Here, since the dance sentences are respectively matched for each phrase in the target music, the matched dance sentences can come from each dance sentence in a plurality of dance sets, and the diversity of dance motion data matched by the target music can be improved.
In the embodiment of the application, the obtained target music of the dance action data to be matched is divided into a plurality of phrases, the phrase matched with each phrase can be determined according to the characteristic information of each phrase, the time sequence information of each phrase in the target music and the pre-stored phrase characteristic information corresponding to each phrase, wherein the phrases matched with any two phrases in the preset time interval range of the target music are different, and the phrases matched with the phrases are spliced according to the time sequence information of each phrase, so that the dance action data matched with the target music can be generated. By adopting the scheme, the diverse dance action data matched with the target music can be quickly generated without model training, the generation efficiency is high, and the matching flexibility is high.
Fig. 2 is a flowchart of a dance action data generation method according to an embodiment of the present application. As shown in fig. 2, a method for generating dance motion data provided in an embodiment of the present application includes the following steps:
s201: and acquiring target music to be matched with the dance action data.
The description of S201 may refer to the description of S101, and the same technical effect can be achieved, which is not described again.
S202: determining a plurality of repetitive music structures in the target music based on the characteristic information of the rhythm point in the target music; each repetitive music structure is determined by the characteristic information of a plurality of consecutive tempo points.
In the specific implementation, to divide the target music into a plurality of phrases, the target music is divided into a plurality of pieces of music, and then the plurality of pieces of music are divided into a plurality of phrases. The target music is divided into a plurality of sections of music, a plurality of repetitive structures in the target music are found out firstly, and then the target music is divided into a plurality of sections of music according to the repetitive structures.
The numbers represented by "plural" in "plural phrases" and "plural pieces" in "plural pieces of music" may be the same or different, and usually "plural" in "plural phrases" is larger than the number represented by "plural pieces" in "plural pieces of music".
Here, the repetitive music structure may be understood as a repetitive music structure between music passages in the target music, for example, a part of music and another part of music in the target music are the same regardless of lyrics, a tune, a tempo and a timbre, and may be considered as music passages having the same repetitive music structure.
Further, in step S202, the determining a plurality of repetitive music structures in the target music based on the characteristic information of the rhythm point in the target music includes the following steps:
calculating a feature vector of a rhythm point in the target music; each element in the feature vector represents the similarity between the rhythm point and each rhythm point in the target music; forming a feature matrix by the feature vectors of all the rhythm points; and determining a plurality of repetitive music structures in the target music based on the feature matrix.
In specific implementation, feature information corresponding to each rhythm point in the target music can be extracted by using the mel frequency spectrum of the target music, and then feature information of each rhythm point is used for calculating a feature vector of each rhythm point, wherein each element in the feature vector of each rhythm point represents the similarity of the mel cepstrum between the rhythm point and each rhythm point in the target music, the feature vectors of each rhythm point of the target music form a feature matrix, the feature matrix can be understood as a recursion matrix of the target music, and each repetitive music structure in the target music can be determined by analyzing the feature matrix.
It should be noted that the mel spectrum and the mel cepstrum are extracted from a very wide range of sound features, the mel spectrum is obtained by processing an audio signal, and the mel cepstrum is obtained by performing cepstrum analysis on the mel spectrum.
S203: and according to the plurality of repetitive music structures, cutting the target music into a plurality of sections of music, and determining the dividing position of a dividing point between any two adjacent sections of music in the plurality of sections of music.
In the specific implementation, after a plurality of repetitive structures in the target music are determined, each repetitive structure is independently used as a music paragraph, the target music is further divided into a plurality of sections of music, after the plurality of sections of music are determined, a music point between any two adjacent sections of music in the plurality of sections of music is used as a dividing point, and the dividing position of each dividing point in the target music is determined.
Further, in step S203, the segmenting the target music into a plurality of pieces of music according to the plurality of repetitive music structures includes the following steps:
clustering the plurality of repetitive music structures to obtain a clustering result; and according to the clustering result, segmenting the target music into the plurality of sections of music.
In specific implementation, the existing clustering algorithm can be utilized to cluster a plurality of repetitive structures in the target music to obtain a clustering structure, so that the target music can be divided into a plurality of sections of music according to the clustering structure.
It should be noted that cluster analysis is also called cluster analysis, and is based on similarity, and the core idea is that there is more similarity between patterns in one cluster than between patterns not in the same cluster, and in general, a clustering algorithm determines clusters based on euclidean or manhattan distance measures. Algorithms based on such distance metrics tend to find spherical clusters with similar dimensions and densities.
S204: and adjusting the segmentation position of the segmentation point according to the beat information of the target music, and adding at least one segmentation compensation point.
In a specific implementation, after the target music is divided into a plurality of pieces of music, the dividing position of the dividing point may be adjusted according to the beat information of the target music, where the adjustment may be understood as moving the position of the dividing point from one position to another position, and since there is a long paragraph in the plurality of pieces of music, if the length of the paragraph exceeds the target music beat, the position of the dividing point needs to be adjusted, and if the length of the paragraph is several times the length of the target music beat, the paragraph needs to be further divided, and in this case, a division compensation point needs to be added.
In one example, the tempo of the target music is 16 beats, and if the tempo of a certain piece of music is 34 beats, a division compensation point may be added between 16 beats and 17 beats, and the position of the division point may be adjusted from between 34 beats and 35 beats to between 32 beats and 33 beats.
S205: and cutting the target music into the plurality of phrases according to the division point after the division position is adjusted and the at least one division compensation point.
In a specific implementation, after the positions of the division points are adjusted and the division compensation points are added, the target music can be divided into a plurality of phrases by using the adjusted division points and the division compensation points.
S206: determining a dance sentence matched with each phrase according to the characteristic information of each phrase in the phrases, the time sequence information of each phrase in the target music and pre-stored dance sentence characteristic information corresponding to the phrases; and the dance sentences matched with any two phrases in the preset time interval range of the target music are different.
The description of S206 may refer to the description of S102, and the same technical effect can be achieved, which is not described in detail herein.
Further, the dance sentences are stored in a dance database in advance; generating the dance database according to the following steps:
obtaining multi-section dances; aiming at each dance in the multi-section dances, dividing each dance into a plurality of dancing sentences according to the characteristic information of each dance action in each dance; and generating the dance database based on the segmented dance sentences and the dance action characteristic information corresponding to each dance sentence.
In the specific implementation, a dance database can be established in advance, and a large number of dance sentences with characteristic information are stored in the dance database, so that dance sentences matched with a plurality of dance sentences in target music can be selected from the dance database when dance motion data are matched with the target music. Here, a large number of existing dances can be acquired, and for an acquired dance segment, each dance segment can be divided into a plurality of dance sentences according to the feature information of the dance segment, so that the divided dance sentences with the feature information are stored to generate a dance database. Here, the feature information of the dance movement includes a movement type, a movement speed, and the movement type is, for example, a rotation, which is further divided into a flat circle, a vertical circle, and the like; the motion speed is uniform, fast, and the like.
Further, the step S206 of determining a dance sentence matched with each phrase according to the feature information of each phrase in the phrases, the timing information of each phrase in the target music, and pre-stored dance sentence feature information respectively corresponding to the phrases, includes the following steps:
aiming at each phrase in the phrases, calculating the matching degree of the phrase and each dance phrase in the phrases according to the characteristic information of the phrase, the time sequence information of the phrase and the pre-stored dance phrase characteristic information corresponding to the phrases; and determining the dance sentences matched with the phrases based on the matching degree of each phrase in the phrases and the dance sentences.
In a specific implementation, when a dance sentence is matched for each phrase, each dance sentence in each dance sentence can be respectively obtained from the dance database, the matching degree between the phrase and the dance sentence can be calculated by using the feature information of the phrase, the time sequence information of the phrase and the obtained feature information of the dance sentence, the matching degree between the phrase and each dance sentence in the dance database can be calculated by using the same method, and the dance sentence with higher or highest matching degree can be selected from the dance sentences as the dance sentence matched with the phrase.
It should be noted that, when calculating the matching degree between a phrase and a dance phrase, a matching degree calculation model may be trained in advance, where the matching degree calculation model may be a markov column model, so that feature information of the phrase, time sequence information of the phrase in target music, and feature information of the dance phrase are input together into the trained matching degree calculation model, and the matching degree between the phrase and the dance phrase may be output.
Further, the determining a dance sentence matched with each phrase based on the matching degree of each phrase in the phrases and the dance sentences respectively comprises the following steps:
step (1): and acquiring a plurality of first candidate matching dancing sentences corresponding to the first phrase in the target music.
In a specific implementation, if the first phrase is the first phrase of the target music, the dance phrases can be matched for the first phrase only according to the matching degree; and if the first phrase is not the first phrase of the target music but is a phrase in a position except the first phrase in the target music, matching the dance phrases for the first phrase according to the matching degree and the dance phrases matched with the phrases before the first phrase. When the first phrase is matched with the dance phrases, a plurality of first candidate matching dance phrases which can be matched with the first phrase can be selected from the dance database preliminarily according to the information, and the dance phrases matched with the first phrase can be determined from the plurality of first candidate matching dance phrases according to the number of second candidate matching dance phrases matched with the second phrase by using a wave function collapse algorithm.
It should be noted that wave function collapse means that when a substance is observed, its state is determined, and this "state is determined" process is called "wave function collapse", and the core principle of the wave function collapse algorithm is to dynamically make the range of candidate objects at each position smaller and smaller until all positions can be selected to be suitable objects.
Step (2): determining a plurality of second candidate matching dances corresponding to a second phrase according to the matching degrees respectively corresponding to the second phrase and the plurality of dances after the first phrase and the incidence relation between each first candidate matching dance in the plurality of first candidate matching dances and the plurality of dances; the first phrase and the second phrase are any adjacent two phrases in the target music.
In a specific implementation, when the second phrase is matched with the dance phrases, a plurality of second candidate matching dance phrases corresponding to the second phrase are determined together according to the matching degree between the second phrase and each dance phrase in the dance database and the association relationship between each dance phrase and each first candidate matching dance phrase in the dance database, wherein the association relationship between the dance phrases can be understood as whether two dance phrases are related or not, whether the two dance phrases are suitable to be connected or not, whether styles are matched or not, and the like.
And (3): and selecting the first candidate matching dancing sentence with the least second candidate matching dancing sentence as the dancing sentence matched with the first phrase from the plurality of first candidate matching dancing sentences.
In the specific implementation, under each first candidate matching dance sentence, a plurality of second candidate matching dance sentences are preliminarily matched for the second phrase, and here, the first candidate matching dance sentences corresponding to the second candidate matching dance sentences with the least number are selected as the dance sentences matched with the first phrases in advance by using the thought of the wave function collapse algorithm.
It should be noted that the first phrase and the second phrase are in a mutually dependent relationship when matching the dance phrase, and here, the first phrase and the second phrase are adjacent to each other. By adopting the matching algorithm, the dance sentences can be matched for each phrase in the target music quickly, and dance motion data matched with the target music can be obtained.
In one example, if there are three first candidate matching dances corresponding to a first phrase, which are respectively the first candidate matching dances A, B, C, the number of second candidate matching dances matching a second phrase is 2 in the case where the first phrase matching dances are the first candidate matching dances a, the number of second candidate matching dances matching the second phrase is 3 in the case where the first phrase matching dances are the first candidate matching dances B, and the number of second candidate matching dances matching the second phrase is 5 in the case where the first phrase matching dances are the first candidate matching dances C, where the number of second candidate matching dances corresponding to the first candidate matching dances a is the minimum (2), the first candidate matching dances a is determined to be the first phrase matching the first phrase.
S207: and splicing the dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases to generate dance motion data matched with the target music.
The description of S207 may refer to the description of S103, and the same technical effect can be achieved, which is not described again.
In the embodiment of the application, the obtained target music of the dance action data to be matched is divided into a plurality of phrases, the phrase matched with each phrase can be determined according to the characteristic information of each phrase, the time sequence information of each phrase in the target music and the pre-stored phrase characteristic information corresponding to each phrase, wherein the phrases matched with any two phrases in the preset time interval range of the target music are different, and the phrases matched with the phrases are spliced according to the time sequence information of each phrase, so that the dance action data matched with the target music can be generated. By adopting the scheme, the diverse dance action data matched with the target music can be quickly generated without model training, the generation efficiency is high, and the matching flexibility is high.
Based on the same application concept, embodiments of the present application further provide a dance action data generation device corresponding to the dance action data generation method embodiment, and as the principle of solving the problem of the device in the embodiments of the present application is similar to the dance action data generation method in the above-described embodiments of the present application, the implementation of the device may refer to the implementation of the method, and repeated details are not described again.
Referring to fig. 3 to 6, fig. 3 is a functional block diagram of a dance motion data generating apparatus 300 according to an embodiment of the present disclosure, fig. 4 is a functional block diagram of an acquiring module 310 in fig. 3, fig. 5 is a second functional block diagram of the dance motion data generating apparatus 300 according to an embodiment of the present disclosure, and fig. 6 is a functional block diagram of a determining module 320 in fig. 3.
As shown in fig. 3 and 5, the dance motion data generating apparatus 300 includes:
an obtaining module 310, configured to obtain target music of dance motion data to be matched, and divide the target music into multiple phrases;
the determining module 320 is configured to determine a dance sentence matched with each phrase according to feature information of each phrase in the phrases, timing information of each phrase in the target music, and pre-stored dance sentence feature information corresponding to each of the phrases; the dance sentences matched with any two phrases in the preset time interval range of the target music are different;
the first generating module 330 is configured to splice dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases, and generate dance motion data matched with the target music.
In a possible implementation manner, as shown in fig. 3 and fig. 5, the obtaining module 310 is specifically configured to obtain the target music according to the following steps:
and acquiring the target music to be matched with the dance action data in a game scene or animation scene of the terminal.
In one possible implementation, as shown in fig. 4, the obtaining module 310 includes:
a first determining unit 312, configured to determine a plurality of repetitive music structures in the target music based on feature information of rhythm points in the target music; each repetitive music structure is determined by the characteristic information of a plurality of continuous rhythm points;
a first dividing unit 314, configured to divide the target music into multiple pieces of music according to the multiple repetitive music structures, and determine a dividing position of a dividing point between any two adjacent pieces of music in the multiple pieces of music;
an adjusting unit 316, configured to adjust a dividing position of the dividing point according to beat information of the target music, and add at least one dividing compensation point;
a second segmenting unit 318, configured to segment the target music into the plurality of phrases according to the segmentation point after the adjustment of the segmentation position and the at least one segmentation compensation point.
In a possible implementation, as shown in fig. 4, the first determining unit 312 is configured to determine the repetitive music structure according to the following steps:
calculating a feature vector of a rhythm point in the target music; each element in the feature vector represents the similarity between the rhythm point and each rhythm point in the target music;
forming a feature matrix by the feature vectors of all the rhythm points;
and determining a plurality of repetitive music structures in the target music based on the feature matrix.
In one possible implementation, as shown in fig. 4, the first dividing unit 314 is configured to divide the target music into a plurality of pieces of music according to the following steps:
clustering the plurality of repetitive music structures to obtain a clustering result;
and according to the clustering result, segmenting the target music into the plurality of sections of music.
In one possible embodiment, as shown in FIG. 5, the plurality of dance sentences are pre-stored in a dance database; the dance action data generation device 300 further comprises a second generation module 340; the second generating module 340 is configured to generate the dance database according to the following steps:
obtaining multi-section dances;
aiming at each dance in the multi-section dances, dividing each dance into a plurality of dancing sentences according to the characteristic information of each dance action in each dance;
and generating the dance database based on the segmented dance sentences and the dance action characteristic information corresponding to each dance sentence.
In one possible implementation, as shown in fig. 6, the determining module 320 includes:
a calculating unit 322, configured to calculate, for each phrase in the multiple phrases, a matching degree between the phrase and each dance phrase in the multiple dance phrases according to feature information of the phrase, time sequence information of the phrase, and pre-stored dance phrase feature information corresponding to the multiple dance phrases;
a second determining unit 324, configured to determine a dance sentence matched with each of the plurality of phrases based on a matching degree of each of the plurality of phrases and the plurality of dance sentences.
In a possible embodiment, as shown in fig. 6, the second determining unit 324 is configured to determine a dance sentence matching each phrase according to the following steps:
obtaining a plurality of first candidate matching dancing sentences corresponding to first phrases in the target music;
determining a plurality of second candidate matching dances corresponding to a second phrase according to the matching degrees respectively corresponding to the second phrase and the plurality of dances after the first phrase and the incidence relation between each first candidate matching dance in the plurality of first candidate matching dances and the plurality of dances; the first phrase and the second phrase are any two adjacent phrases in the target music;
and selecting the first candidate matching dancing sentence with the least second candidate matching dancing sentence as the dancing sentence matched with the first phrase from the plurality of first candidate matching dancing sentences.
In the embodiment of the application, the obtained target music of the dance action data to be matched is divided into a plurality of phrases, the phrase matched with each phrase can be determined according to the characteristic information of each phrase, the time sequence information of each phrase in the target music and the pre-stored phrase characteristic information corresponding to each phrase, wherein the phrases matched with any two phrases in the preset time interval range of the target music are different, and the phrases matched with the phrases are spliced according to the time sequence information of each phrase, so that the dance action data matched with the target music can be generated. By adopting the scheme, the diverse dance action data matched with the target music can be quickly generated without model training, the generation efficiency is high, and the matching flexibility is high.
Based on the same application concept, referring to fig. 7, a schematic structural diagram of an electronic device 700 provided in the embodiment of the present application includes: a processor 710, a memory 720 and a bus 730, wherein the memory 720 stores machine-readable instructions executable by the processor 710, when the electronic device 700 is operated, the processor 710 communicates with the memory 720 through the bus 730, and the machine-readable instructions are executed by the processor 710 to perform the steps of the dance motion data generation method as described in the embodiment.
In particular, the machine readable instructions, when executed by the processor 710, may perform the following:
obtaining target music of dance action data to be matched, and dividing the target music into a plurality of phrases;
determining a dance sentence matched with each phrase according to the characteristic information of each phrase in the phrases, the time sequence information of each phrase in the target music and pre-stored dance sentence characteristic information corresponding to the phrases; the dance sentences matched with any two phrases in the preset time interval range of the target music are different;
and splicing the dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases to generate dance motion data matched with the target music.
In the embodiment of the application, the obtained target music of the dance action data to be matched is divided into a plurality of phrases, the phrase matched with each phrase can be determined according to the characteristic information of each phrase, the time sequence information of each phrase in the target music and the pre-stored phrase characteristic information corresponding to each phrase, wherein the phrases matched with any two phrases in the preset time interval range of the target music are different, and the phrases matched with the phrases are spliced according to the time sequence information of each phrase, so that the dance action data matched with the target music can be generated. By adopting the scheme, the diverse dance action data matched with the target music can be quickly generated without model training.
Based on the same application concept, the embodiment of the application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to perform the steps of the dance motion data generation method provided by the above embodiment.
Specifically, the storage medium can be a general-purpose storage medium, such as a removable disk, a hard disk, or the like, and when a computer program on the storage medium is executed, the method for generating dance motion data can be executed, and diverse dance motion data matched with target music can be generated quickly without performing model training.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (11)
1. A dance action data generation method, comprising:
obtaining target music of dance action data to be matched, and dividing the target music into a plurality of phrases;
determining a dance sentence matched with each phrase according to the characteristic information of each phrase in the phrases, the time sequence information of each phrase in the target music and pre-stored dance sentence characteristic information corresponding to the phrases; the dance sentences matched with any two phrases in the preset time interval range of the target music are different;
and splicing the dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases to generate dance motion data matched with the target music.
2. The generation method according to claim 1, wherein the acquiring of the target music of the dance action data to be matched includes:
and acquiring the target music to be matched with the dance action data in a game scene or animation scene of the terminal.
3. The generation method according to claim 1, wherein the target music is sliced into a plurality of phrases according to the following steps:
determining a plurality of repetitive music structures in the target music based on the characteristic information of the rhythm point in the target music; each repetitive music structure is determined by the characteristic information of a plurality of continuous rhythm points;
according to the multiple repetitive music structures, the target music is cut into multiple sections of music, and the dividing position of a dividing point between any two adjacent sections of music in the multiple sections of music is determined;
adjusting the dividing positions of the dividing points according to the beat information of the target music, and adding at least one dividing compensation point;
and cutting the target music into the plurality of phrases according to the division point after the division position is adjusted and the at least one division compensation point.
4. The generation method according to claim 3, wherein the determining a plurality of repetitive music structures in the target music based on the feature information of the rhythm point in the target music comprises:
calculating a feature vector of a rhythm point in the target music; each element in the feature vector represents the similarity between the rhythm point and each rhythm point in the target music;
forming a feature matrix by the feature vectors of all the rhythm points;
and determining a plurality of repetitive music structures in the target music based on the feature matrix.
5. The method of generating as claimed in claim 3, wherein said segmenting said target music into a plurality of pieces of music according to said plurality of repeating music structures comprises:
clustering the plurality of repetitive music structures to obtain a clustering result;
and according to the clustering result, segmenting the target music into the plurality of sections of music.
6. The generation method according to claim 1, wherein the plurality of dance sentences are stored in a dance database in advance; the generating method further comprises generating the dance database according to the following steps:
obtaining multi-section dances;
aiming at each dance in the multi-section dances, dividing each dance into a plurality of dancing sentences according to the characteristic information of each dance action in each dance;
and generating the dance database based on the segmented dance sentences and the dance action characteristic information corresponding to each dance sentence.
7. The generating method of claim 1, wherein the determining the dance sentences matched with each phrase according to the feature information of each phrase in the phrases, the time sequence information of each phrase in the target music, and pre-stored dance sentence feature information corresponding to each phrase comprises:
aiming at each phrase in the phrases, calculating the matching degree of the phrase and each dance phrase in the phrases according to the characteristic information of the phrase, the time sequence information of the phrase and the pre-stored dance phrase characteristic information corresponding to the phrases;
and determining the dance sentences matched with the phrases based on the matching degree of each phrase in the phrases and the dance sentences.
8. The generating method of claim 7, wherein determining the dance sentence matched with each of the plurality of phrases based on the matching degree of each of the plurality of phrases with the plurality of dance sentences comprises:
obtaining a plurality of first candidate matching dancing sentences corresponding to first phrases in the target music;
determining a plurality of second candidate matching dances corresponding to a second phrase according to the matching degrees respectively corresponding to the second phrase and the plurality of dances after the first phrase and the incidence relation between each first candidate matching dance in the plurality of first candidate matching dances and the plurality of dances; the first phrase and the second phrase are any two adjacent phrases in the target music;
and selecting the first candidate matching dancing sentence with the least second candidate matching dancing sentence as the dancing sentence matched with the first phrase from the plurality of first candidate matching dancing sentences.
9. A dance motion data generation apparatus, comprising:
the dance action data matching module is used for matching dance action data with dance action data to obtain target music;
the determining module is used for determining a dance sentence matched with each phrase according to the characteristic information of each phrase in the phrases, the time sequence information of each phrase in the target music and pre-stored dance sentence characteristic information corresponding to each phrase; the dance sentences matched with any two phrases in the preset time interval range of the target music are different;
and the first generating module is used for splicing the dance sentences respectively matched with the multiple phrases according to the time sequence information of each phrase in the multiple phrases to generate dance action data matched with the target music.
10. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the electronic device is operating, the machine-readable instructions being executable by the processor to perform the steps of the dance motion data generating method according to any one of claims 1 to 8.
11. A computer-readable storage medium, having stored thereon a computer program which, when executed by a processor, performs the steps of the dance motion data generation method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911206883.0A CN110955786B (en) | 2019-11-29 | 2019-11-29 | Dance action data generation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911206883.0A CN110955786B (en) | 2019-11-29 | 2019-11-29 | Dance action data generation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110955786A true CN110955786A (en) | 2020-04-03 |
| CN110955786B CN110955786B (en) | 2023-10-27 |
Family
ID=69979288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911206883.0A Active CN110955786B (en) | 2019-11-29 | 2019-11-29 | Dance action data generation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110955786B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111798549A (en) * | 2020-07-15 | 2020-10-20 | 广州微咔世纪信息科技有限公司 | Dance editing method and device and computer storage medium |
| CN111970536A (en) * | 2020-07-24 | 2020-11-20 | 北京航空航天大学 | Method and device for generating video based on audio |
| CN111968202A (en) * | 2020-08-21 | 2020-11-20 | 北京中科深智科技有限公司 | Real-time dance action generation method and system based on music rhythm |
| CN112365568A (en) * | 2020-11-06 | 2021-02-12 | 广州小鹏汽车科技有限公司 | Audio processing method and device, electronic equipment and storage medium |
| CN112700521A (en) * | 2021-01-26 | 2021-04-23 | 上海交通大学 | Music-driven human skeleton dance motion generation system |
| CN112894855A (en) * | 2021-02-23 | 2021-06-04 | 广东智源机器人科技有限公司 | Robot motion generation method and device, robot, and storage medium |
| CN113521711A (en) * | 2021-07-13 | 2021-10-22 | 济南幼儿师范高等专科学校 | Dance training auxiliary system and method |
| CN113643442A (en) * | 2021-08-06 | 2021-11-12 | 福建凯米网络科技有限公司 | Virtual object dance generation method and computer storage medium |
| CN114741561A (en) * | 2022-02-28 | 2022-07-12 | 商汤国际私人有限公司 | Action generating method, device, electronic equipment and storage medium |
| CN114827679A (en) * | 2022-04-19 | 2022-07-29 | 海信视像科技股份有限公司 | Display device and sound picture synchronization method |
| CN115379299A (en) * | 2022-08-23 | 2022-11-22 | 清华大学 | Dance action generation method and device, electronic equipment and storage medium |
| CN115712739A (en) * | 2022-11-17 | 2023-02-24 | 腾讯音乐娱乐科技（深圳）有限公司 | Dance action generation method, computer device and storage medium |
| CN115861492A (en) * | 2022-12-15 | 2023-03-28 | 北京百度网讯科技有限公司 | Audio-based action generation method, device, equipment and medium |
| WO2023115925A1 (en) * | 2021-12-22 | 2023-06-29 | 北京百度网讯科技有限公司 | Virtual digital human driving method and pose acquisition model training method |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5315057A (en) * | 1991-11-25 | 1994-05-24 | Lucasarts Entertainment Company | Method and apparatus for dynamically composing music and sound effects using a computer entertainment system |
| WO2007096691A2 (en) * | 2006-02-21 | 2007-08-30 | Nokia Corporation | Generating a representation of a dancer dancing to music |
| JP2008065153A (en) * | 2006-09-08 | 2008-03-21 | Fujifilm Corp | Musical piece structure analyzing method, program and device |
| CN101524594A (en) * | 2009-04-17 | 2009-09-09 | 中国地质大学（武汉） | Anthropomorphic robot autonomously dancing along with rhythm |
| CN101615302A (en) * | 2009-07-30 | 2009-12-30 | 浙江大学 | The dance movement generation method that music data drives based on machine learning |
| WO2010057537A1 (en) * | 2008-11-24 | 2010-05-27 | Movea | System for computer-assisted interpretation of pre-recorded music |
| CN103793446A (en) * | 2012-10-29 | 2014-05-14 | 汤晓鸥 | Music video generation method and system |
| JP2016041142A (en) * | 2014-08-15 | 2016-03-31 | 国立研究開発法人産業技術総合研究所 | Dance motion data creation system and dance motion data creation method |
| CN105612510A (en) * | 2013-08-28 | 2016-05-25 | 兰德音频有限公司 | System and method for performing automatic audio production using semantic data |
| CN107483843A (en) * | 2017-08-16 | 2017-12-15 | 成都品果科技有限公司 | Audio frequency and video match clipping method and device |
| CN108202334A (en) * | 2018-03-22 | 2018-06-26 | 东华大学 | A kind of Dancing Robot that can identify music beat and style |
| CN108527376A (en) * | 2018-02-27 | 2018-09-14 | 深圳狗尾草智能科技有限公司 | Control method, apparatus, equipment and the medium of robot dance movement |
| CN108665492A (en) * | 2018-03-27 | 2018-10-16 | 北京光年无限科技有限公司 | A kind of Dancing Teaching data processing method and system based on visual human |
| CN109144610A (en) * | 2018-08-31 | 2019-01-04 | 腾讯科技（深圳）有限公司 | Audio frequency playing method, device, electronic device and computer readable storage medium |
| CN109176541A (en) * | 2018-09-06 | 2019-01-11 | 南京阿凡达机器人科技有限公司 | A kind of method, equipment and storage medium realizing robot and dancing |
-
2019
- 2019-11-29 CN CN201911206883.0A patent/CN110955786B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5315057A (en) * | 1991-11-25 | 1994-05-24 | Lucasarts Entertainment Company | Method and apparatus for dynamically composing music and sound effects using a computer entertainment system |
| WO2007096691A2 (en) * | 2006-02-21 | 2007-08-30 | Nokia Corporation | Generating a representation of a dancer dancing to music |
| JP2008065153A (en) * | 2006-09-08 | 2008-03-21 | Fujifilm Corp | Musical piece structure analyzing method, program and device |
| WO2010057537A1 (en) * | 2008-11-24 | 2010-05-27 | Movea | System for computer-assisted interpretation of pre-recorded music |
| CN101524594A (en) * | 2009-04-17 | 2009-09-09 | 中国地质大学（武汉） | Anthropomorphic robot autonomously dancing along with rhythm |
| CN101615302A (en) * | 2009-07-30 | 2009-12-30 | 浙江大学 | The dance movement generation method that music data drives based on machine learning |
| CN103793446A (en) * | 2012-10-29 | 2014-05-14 | 汤晓鸥 | Music video generation method and system |
| CN105612510A (en) * | 2013-08-28 | 2016-05-25 | 兰德音频有限公司 | System and method for performing automatic audio production using semantic data |
| JP2016041142A (en) * | 2014-08-15 | 2016-03-31 | 国立研究開発法人産業技術総合研究所 | Dance motion data creation system and dance motion data creation method |
| CN107483843A (en) * | 2017-08-16 | 2017-12-15 | 成都品果科技有限公司 | Audio frequency and video match clipping method and device |
| CN108527376A (en) * | 2018-02-27 | 2018-09-14 | 深圳狗尾草智能科技有限公司 | Control method, apparatus, equipment and the medium of robot dance movement |
| CN108202334A (en) * | 2018-03-22 | 2018-06-26 | 东华大学 | A kind of Dancing Robot that can identify music beat and style |
| CN108665492A (en) * | 2018-03-27 | 2018-10-16 | 北京光年无限科技有限公司 | A kind of Dancing Teaching data processing method and system based on visual human |
| CN109144610A (en) * | 2018-08-31 | 2019-01-04 | 腾讯科技（深圳）有限公司 | Audio frequency playing method, device, electronic device and computer readable storage medium |
| CN109176541A (en) * | 2018-09-06 | 2019-01-11 | 南京阿凡达机器人科技有限公司 | A kind of method, equipment and storage medium realizing robot and dancing |
Non-Patent Citations (7)
| Title |
|---|
| 方丹芳: "基于过渡帧插值的音乐驱动舞蹈动作合成", 《复旦学报(自然科学版)》 * |
| 方丹芳: "基于过渡帧插值的音乐驱动舞蹈动作合成", 《复旦学报(自然科学版)》, 30 June 2018 (2018-06-30) * |
| 方丹芳: "音乐驱动的舞蹈动作合成", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 10, pages 5 - 6 * |
| 樊儒昆: "音乐驱动的舞蹈动作合成", 《中国优秀博硕士学位论文全文数据库（硕士）信息科技辑》 * |
| 樊儒昆: "音乐驱动的舞蹈动作合成", 《中国优秀博硕士学位论文全文数据库（硕士）信息科技辑》, 15 August 2010 (2010-08-15), pages 7 - 20 * |
| 石祥滨: "一种分段式音乐情感识别方法", 《小型微型计算机系统》 * |
| 石祥滨: "一种分段式音乐情感识别方法", 《小型微型计算机系统》, 15 August 2012 (2012-08-15) * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111798549B (en) * | 2020-07-15 | 2024-02-20 | 广州微咔世纪信息科技有限公司 | Dance editing method and device and computer storage medium |
| CN111798549A (en) * | 2020-07-15 | 2020-10-20 | 广州微咔世纪信息科技有限公司 | Dance editing method and device and computer storage medium |
| CN111970536A (en) * | 2020-07-24 | 2020-11-20 | 北京航空航天大学 | Method and device for generating video based on audio |
| CN111970536B (en) * | 2020-07-24 | 2021-07-23 | 北京航空航天大学 | Method and device for generating video based on audio |
| CN111968202A (en) * | 2020-08-21 | 2020-11-20 | 北京中科深智科技有限公司 | Real-time dance action generation method and system based on music rhythm |
| CN112365568A (en) * | 2020-11-06 | 2021-02-12 | 广州小鹏汽车科技有限公司 | Audio processing method and device, electronic equipment and storage medium |
| CN112700521A (en) * | 2021-01-26 | 2021-04-23 | 上海交通大学 | Music-driven human skeleton dance motion generation system |
| CN112894855A (en) * | 2021-02-23 | 2021-06-04 | 广东智源机器人科技有限公司 | Robot motion generation method and device, robot, and storage medium |
| CN113521711A (en) * | 2021-07-13 | 2021-10-22 | 济南幼儿师范高等专科学校 | Dance training auxiliary system and method |
| CN113643442A (en) * | 2021-08-06 | 2021-11-12 | 福建凯米网络科技有限公司 | Virtual object dance generation method and computer storage medium |
| CN113643442B (en) * | 2021-08-06 | 2024-05-14 | 福建凯米网络科技有限公司 | Virtual object dance generation method and computer storage medium |
| WO2023115925A1 (en) * | 2021-12-22 | 2023-06-29 | 北京百度网讯科技有限公司 | Virtual digital human driving method and pose acquisition model training method |
| CN114741561A (en) * | 2022-02-28 | 2022-07-12 | 商汤国际私人有限公司 | Action generating method, device, electronic equipment and storage medium |
| CN114827679A (en) * | 2022-04-19 | 2022-07-29 | 海信视像科技股份有限公司 | Display device and sound picture synchronization method |
| CN115379299A (en) * | 2022-08-23 | 2022-11-22 | 清华大学 | Dance action generation method and device, electronic equipment and storage medium |
| CN115712739A (en) * | 2022-11-17 | 2023-02-24 | 腾讯音乐娱乐科技（深圳）有限公司 | Dance action generation method, computer device and storage medium |
| CN115712739B (en) * | 2022-11-17 | 2024-03-26 | 腾讯音乐娱乐科技（深圳）有限公司 | Dance motion generation method, computer device and storage medium |
| CN115861492A (en) * | 2022-12-15 | 2023-03-28 | 北京百度网讯科技有限公司 | Audio-based action generation method, device, equipment and medium |
| CN115861492B (en) * | 2022-12-15 | 2023-09-29 | 北京百度网讯科技有限公司 | Audio-based action generation method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110955786B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110955786B (en) | Dance action data generation method and device | |
| CN108305643B (en) | Method and device for determining emotion information | |
| CN110264991A (en) | Training method, phoneme synthesizing method, device, equipment and the storage medium of speech synthesis model | |
| WO2019232928A1 (en) | Musical model training method, music creation method, devices, terminal and storage medium | |
| JP2017513049A (en) | How to provide users with feedback on the performance of karaoke songs | |
| CN112542155B (en) | Song synthesis method, model training method, device, equipment and storage medium | |
| CN111127598B (en) | Animation playing speed adjusting method and device, electronic equipment and medium | |
| CN111179385B (en) | Dance animation processing method and device, electronic equipment and storage medium | |
| CN111354332A (en) | Singing voice synthesis method and device | |
| CN109326280B (en) | Singing synthesis method and device and electronic equipment | |
| CN113010730B (en) | Music file generation method, device, equipment and storage medium | |
| CN113160848A (en) | Dance animation generation method, dance animation model training method, dance animation generation device, dance animation model training device, dance animation equipment and storage medium | |
| JP2020003535A (en) | Program, information processing method, electronic apparatus and learnt model | |
| Fukayama et al. | Music content driven automated choreography with beat-wise motion connectivity constraints | |
| CN113823323A (en) | Audio processing method and device based on convolutional neural network and related equipment | |
| CN114449313B (en) | Method and device for adjusting audio and video playing rate of video | |
| CN113920968A (en) | Information processing method, information processing device, electronic equipment and storage medium | |
| CN113158642A (en) | Information processing method, information processing device, electronic equipment and storage medium | |
| CN113140230A (en) | Method, device and equipment for determining pitch value of note and storage medium | |
| US11081138B2 (en) | Systems and methods for automated music rearrangement | |
| CN110708619B (en) | Word vector training method and device for intelligent equipment | |
| CN111863030A (en) | Audio detection method and device | |
| CN115375806A (en) | Dance motion prediction model training method, dance synthesis equipment and dance motion prediction model product | |
| CN111339350B (en) | Data processing method and device, storage medium and electronic equipment | |
| CN114820888A (en) | Animation generation method and system and computer equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |