A digital display system for ethnic dances

Abstract

The present invention relates to the technical field of digital display, and specifically provides a digital display system for ethnic dances. The system includes a trajectory node extraction module, a rhythm segment optimization module, a trend stability recognition module, an action information acquisition module, and a synchronous recombination execution module. In the present invention, by constructing a spatial vector to extract angle mutation points, the recognition of action changes is strengthened, and the accuracy of frame sequence structure analysis is improved. The screening of rhythm paragraphs introduces a double stability criterion of duration and amplitude, enhancing the accuracy and robustness of rhythm recognition. By comparing the slope fluctuations to extract trend-stable frames, the continuity of the rhythm flow and the centrality of node selection are ensured. The main limb movements are constructed based on the total path distance, direction angle, and frame time sequence, ensuring clear structure and consistent direction of action expression. During the dynamic recombination process, a direction correction and beat scaling strategy are introduced to ensure smooth action transition and rhythm synchronization, significantly improving the rhythm coordination and dynamic coherence of dance movements.

Description

Technical Field

[0001] The present invention relates to the technical field of digital display, and in particular to a digital display system for ethnic dances. Background Art

[0002] The technical field of digital display includes relevant methods and applications for collecting information, digital modeling, visual processing, and interactive presentation of physical or cultural objects using digital media and computer technology. The core contents include image acquisition and processing, three-dimensional modeling, digital content reconstruction, visual interaction design, and display terminal presentation, etc. Digital display is widely used in multiple directions such as cultural heritage protection, education and training, virtual museums, stage art dissemination, etc. Its overall technical system emphasizes collecting, reconstructing objects or activities in the real world in digital form with high fidelity and presenting them visually to users, realizing digital reproduction and interactive expression of physical spaces or cultural contents.

[0003] Among them, the digital display system for ethnic dances refers to a dedicated digital platform for collecting, processing, and displaying ethnic dance cultural resources, covering the acquisition of dance movement data, three-dimensional modeling of dance forms, digital animation production during the dance performance process, and graphical display methods of dance contents. Specifically, it dynamically collects the body movements of dancers using three-dimensional motion capture technology, generates an editable three-dimensional dance movement model in combination with computer graphics processing methods, and converts the model movements into visual images through graphical rendering means for digital demonstration on display terminal devices.

[0004] In the existing ethnic dance display process, there is a lack of multi-dimensional analysis of the dynamic structure in the dance movement sequence during the operation process. It often relies on modeling and rendering the overall movement, and fails to deeply extract the internal change characteristics and rhythm rules of the movement, resulting in a relatively flat presentation content. The method for identifying key nodes in the frame sequence is single, relying only on continuous movement records rather than extracting based on spatial change characteristics, causing inaccurate judgment of the movement undulation between frames and affecting the structural accuracy of subsequent reconstruction. At the rhythm processing level, there is no stable screening mechanism for identifying different beat paragraphs, and the beat undulation is not effectively distinguished, resulting in problems such as blurred rhythm expression and chaotic beats. In terms of trend processing, the change trend of the movement trend is ignored, and the movement is only presented in the form of a surface time series, lacking the recognition of the internal change trend and affecting the temporal tension of the display content. The display content mainly focuses on the reproduction of original movements, without achieving unified direction and beat reconstruction, resulting in a lack of overall rhythm and structural logic in the dynamic expression form. For example, when displaying dance segments with obvious rhythm undulations, high-frequency oscillations or direction jumps are likely to occur, interfering with the accuracy and appreciation of dance language. Summary of the Invention

[0005] The object of the present invention is to solve the disadvantages existing in the prior art, and to propose a digital display system for ethnic dances.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A digital display system for ethnic dances includes:

[0007] The trajectory node extraction module obtains the three-dimensional position coordinates of ethnic dance movements, constructs spatial vectors and calculates the included angle values, judges and records the angle mutation points, and generates a turning frame index sequence;

[0008] The rhythm segment optimization module, based on the turning frame index sequence, obtains the beat duration and beat amplitude of the turning paragraph, calculates the maximum and minimum differences of the beat duration and amplitude respectively, compares with the beat stability threshold, and filters out the beat stable paragraphs;

[0009] The trend stability recognition module, according to the beat stable paragraphs, extracts the spatial position coordinates of the corresponding frame points, performs linear fitting and extracts the corresponding fitting slope, judges the continuous paragraph with the smallest fluctuation amplitude and extracts the frame index, and generates the trend stable node index data;

[0010] The action information acquisition module, according to the trend stable node index data, obtains the three-dimensional position points of each frame of the main limb, calculates the total path distance, constructs a direction vector and calculates the direction angle, and combines the starting frame and the total number of frames to obtain the main limb action information;

[0011] The synchronization and reorganization execution module, according to the main limb action information, calculates the rotation angle and performs direction correction, scales the time duration values of all frames proportionally, rearranges the frame sequence to construct a new dynamic structure, and obtains the dynamic display result of the ethnic dance.

[0012] As a further solution of the present invention, the turning frame index sequence includes action change nodes, turning angle distributions, action segmentation boundaries, the beat stable paragraphs include beat balance indexes, rhythm start and end information, stability interval identifiers, the trend stable node index data includes stable trend reference frames, fitting change ranges, rhythm trend parameters, the main limb action information includes action path features, direction change parameters, action timing structures, and the dynamic display result of the ethnic dance includes a reorganized action frame sequence, a unified direction and posture, and a beat synchronization rhythm.

[0013] As a further solution of the present invention, the trajectory node extraction module includes:

[0014] The three-dimensional coordinate extraction sub-module, based on the video frame sequence of the continuous stepping movements of ethnic dances, obtains the three-dimensional position coordinate data of the bones of the dancer in each frame, sequentially extracts the three-dimensional coordinates of the dancer's feet joints in each frame, organizes and stores them as a three-dimensional coordinate set sorted by frame order, and generates a three-dimensional coordinate sequence of the dance gait;

[0015] The spatial vector construction sub-module constructs two sets of spatial vectors according to the three-dimensional coordinate sequence of the dance gait, calculates the included angle values within each set of three frames, integrates all the included angle results between three frames, and generates a spatial gait included angle value sequence;

[0016] The angle mutation judgment sub-module uses the formula according to the spatial gait included angle value sequence:

[0017] ;

[0018] Calculate the included angle value of the th frame , extract and record the frame positions according to the frame indices where the angle mutation amplitude value is greater than the angle mutation threshold, obtain and establish a turning frame index sequence, where is the spatial vector constructed for the th frame, is the modulus length, is the distance between the two feet key points of the th frame, is the time stamp of the th frame, is the angle mutation amplitude value of the th frame, is the angle mutation amplitude value of the th frame, is the distance between the two feet key points of the th frame, is the time stamp of the th frame.

[0019] As a further solution of the present invention, the rhythm segment optimization module includes:

[0020] The beat feature extraction sub-module extracts the time interval between adjacent turning frames based on the turning frame index sequence to obtain the beat duration, calls the three-dimensional coordinate information within the corresponding frames to calculate the beat amplitude, and generates a rhythm segment beat feature sequence;

[0021] The beat difference calculation sub-module calculates the difference between the maximum value and the minimum value of the beat duration set and the amplitude set within each segment according to the rhythm segment beat feature sequence, using the formula:

[0022] ;

[0023] Calculate the beat difference combination value of the th segment to obtain a beat difference combination value sequence, where , respectively represent the maximum value and the minimum value of the beat duration of the th segment, , respectively represent the maximum and minimum values of the amplitude of the section beat;

[0024] The stable section screening sub-module determines whether the combined value in each rhythm section is less than the beat stability threshold according to the beat difference combined value sequence, screens all rhythm sections that meet the conditions, and establishes a beat stable section.

[0025] As a further solution of the present invention, the trend stability recognition module includes:

[0026] The coordinate extraction sub-module extracts all frame indexes corresponding to each section according to the beat stable section, obtains the three-dimensional space position coordinates corresponding to the frame indexes, combines and records the coordinate sequences of each section in chronological order, and generates stable section space coordinate data;

[0027] The trend fluctuation calculation sub-module performs a linear fitting operation on the sequence of the three-dimensional space coordinates changing with time in each section based on the stable section space coordinate data, obtains the linear fitting slope value in the main time axis direction of each section, compares the change intensity between all slope values, establishes a composite calculation relationship for the slope change amount and the section length between adjacent sections, and uses the formula:

[0028] ;

[0029] Calculate the combined trend fluctuation value , extract the stable trend segment according to the continuous segment number interval with the smallest fluctuation value, and obtain the trend stable interval, where is the linear fitting slope value of the section, is the linear fitting slope value of the section, is the number of frames of the section, is the cumulative operation from the 1st section to the section, and the denominator term is used to avoid division by zero, is the absolute value of the product of the slope value and the inverse section length, and is used to measure the sensitive response of trend instability in a short section, represents the total number of sections;

[0030] The node index generation sub-module extracts all frame index sets in each stable section according to the trend stable interval, calculates the median position according to the time position, extracts the corresponding frame index, and establishes the trend stable node index data.

[0031] As a further solution of the present invention, the action information acquisition module includes:

[0032] The main limb coordinate extraction sub-module extracts all frame indexes of the rhythm segment to which the node belongs according to the trend-stable node index data, obtains the three-dimensional position coordinates of the main limb in each frame, organizes them into a continuous coordinate sequence arranged in frame order, and generates the main limb time-series coordinate sequence;

[0033] Based on the main limb time-series coordinate sequence, the action path calculation sub-module sequentially calculates the spatial distances between adjacent frame points frame by frame, extracts the coordinate difference between the first frame and the last frame to represent the direction, and uses the formula:

[0034] ;

[0035] Calculate the path direction combination value , revealing the total amount of action displacement and direction change, and obtaining the main limb spatial motion parameters. Among them, is the frame and the frame, , are the coordinate values of the main limb on the x-axis of the first frame and the last frame, is the total number of frames in the rhythm segment;

[0036] The action information construction sub-module constructs corresponding structured fields and summarizes them according to the main limb spatial motion parameters, combined with the starting frame number and the number of frames of each segment, and establishes the main limb action information.

[0037] As a further solution of the present invention, the synchronization and reorganization execution module includes:

[0038] The direction correction sub-module traverses all frame points in the current rhythm segment frame by frame according to the main limb action information, extracts the spatial direction vector of the main limb in each frame, calculates the angle difference between the current direction and the starting direction, and adjusts the current frame direction vector to be consistent with the initial direction, generating a direction-consistent vector sequence;

[0039] The time scaling sub-module performs an equal-proportion scaling operation on the time duration value corresponding to each frame based on the direction-consistent vector sequence, and reallocates the new time point position corresponding to each frame according to the actual time lengths of the rhythm segment start time and the beat period, obtaining an equal-proportion alignment time sequence;

[0040] The structure rearrangement sub-module rearranges the order of all frames in the rhythm segment according to the equal-proportion alignment time sequence, and simultaneously reconstructs the coordinates, directions, and time tags corresponding to each frame, and uniformly organizes them into continuous dynamic structure data, obtaining the ethnic dance dynamic display result.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0042] In the present invention, by constructing spatial vectors for every three frames of dance movements and extracting angle mutation points, a dynamic node screening mechanism based on angle features is formed, enhancing the sensitive recognition ability of action changes. The screening of rhythm segments no longer relies solely on temporal segmentation, but comprehensively considers the rhythm duration and amplitude stability differences to accurately lock in the action intervals with clear rhythm structures, improving the discrimination accuracy of rhythm segment selection. The quantitative processing of the fluctuations in the slopes of adjacent linear fittings strengthens the ability to recognize trend continuity, making the node selection have stronger temporal centrality and trend representativeness. In the extraction of the main limb movement features, the three indicators of total path distance, direction angle, and frame time sequence act together to improve the structural integrity and direction consistency of action expression. Subsequently, in the dynamic frame sequence recombination, the methods of rotation angle correction and time line equal-proportion scaling are fused to make the rhythm transition between frames more natural, synchronizing the temporal rhythm and action rhythm, and enhancing the expressiveness of the overall display in terms of dynamic aesthetics and rhythm. The overall processing path starts with structural feature extraction, takes rhythm stability recognition as the axis, and action dynamic recombination as the core, forming a dance digital display framework integrating spatial nodes, rhythm structures, and time regulation, and constructing a more rhythmically coordinated and action-coherent dynamic presentation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is the system flowchart of the present invention;

[0044] Figure 2 is the flowchart of the trajectory node extraction module of the present invention;

[0045] Figure 3 is the flowchart of the rhythm segment optimization module of the present invention;

[0046] Figure 4 is the flowchart of the trend stability recognition module of the present invention;

[0047] Figure 5 is the flowchart of the action information acquisition module of the present invention;

[0048] Figure 6 is the flowchart of the synchronous recombination execution module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0050] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0051] Please refer to Figure 1 , a digital display system for ethnic dances includes:

[0052] The trajectory node extraction module obtains the three-dimensional position coordinates of each frame in the continuous stepping movements of ethnic dances, constructs the spatial vectors between every three frames and calculates the included angle values, makes angle judgments on all the included angle values and records the frame indices of the angle mutation points, and generates a turning frame index sequence;

[0053] The rhythm section optimization module, based on the turning frame index sequence, obtains the beat duration and beat amplitude within each turning paragraph, calculates the difference between the maximum and minimum values of the beat duration and the difference between the maximum and minimum values of the amplitude for each section, compares with the beat stability threshold, and screens out the beat sections that simultaneously satisfy the condition that both differences are less than the stability threshold to obtain the beat stable paragraphs;

[0054] The trend stability recognition module, according to the beat stable paragraphs, extracts the spatial position coordinates of the corresponding frame points, performs linear fitting and extracts the corresponding fitting slopes, compares the fluctuation values between adjacent slopes, determines the continuous paragraph with the smallest fluctuation amplitude and extracts the frame index at the middle position in time in the corresponding paragraph to generate the trend stable node index data;

[0055] The motion information acquisition module, according to the trend stable node index data, obtains the three-dimensional position points of each frame of the main limb in the corresponding rhythm section, calculates the total path distance between consecutive frames, extracts the head and tail frame coordinates to construct a direction vector and calculates the direction angle, and combines the starting frame and the total number of frames to construct the main limb motion information.

[0056] The synchronization and recombination execution module, according to the main limb motion information, traverses all the frame points in the current rhythm section frame by frame, calculates the rotation angle of the direction vector of each frame and performs direction correction, scales the time duration values of all frames proportionally according to the actual timeline of the beat period of the current section, rearranges the frame sequence to construct a new dynamic structure, and obtains the dynamic display result of ethnic dances.

[0057] The turning frame index sequence includes action change nodes, turning angle distribution, and action segmentation boundaries. The beat stable paragraph includes beat balance indicators, rhythm start and end information, and stability interval identification. The trend stable node index data includes stable trend reference frames, fitting change ranges, and rhythm trend parameters. The main limb action information includes action path characteristics, direction change parameters, and action timing structures. The ethnic dance dynamic display results include recombined action frame sequences, unified direction postures, and beat-synchronized rhythms.

[0058] Please refer to Figure 2 , and the trajectory node extraction module includes:

[0059] The three-dimensional coordinate extraction sub-module is based on the ethnic dance continuous stepping action video frame sequence, obtains the three-dimensional position coordinate data of the dancer's bones in each frame, extracts the three-dimensional coordinates of the dancer's feet joints in each frame in turn, organizes and stores them as a three-dimensional coordinate set sorted by frame order, and generates a dance gait three-dimensional coordinate sequence;

[0060] Based on the ethnic dance continuous stepping action video frame sequence, it is necessary to first collect the image information of each frame in the video and locate the key point positions of the dancer in the image. Subsequently, combined with the image recognition processing program, extract the three-dimensional coordinate information of the dancer's feet joints in each frame. In the actual collection process, the shooting frequency can be set to 30 frames per second. After recording 1 minute of dance actions, 1800 frame image data can be obtained. Extract the three-dimensional coordinates of the two feet key points from the image to form a group of three-dimensional vector arrays, where the x, y, and z axis data respectively represent the spatial position coordinates. Traverse each frame of data in turn and sort them in chronological order to obtain a complete coordinate sequence. For example, the left foot coordinate of the 28th frame is (21.3, 48.6, 18.9), and the right foot coordinate is (26.7, 47.9, 19.4). Take the midpoint position of the left and right feet as the gait center point. Through multi-frame coordinate data, a dance gait three-dimensional coordinate sequence can be formed; for the continuous stepping situation in the dance, a frame segment with obvious continuous gait changes can be intercepted for feature extraction operations. During the extraction process, data segments with almost no change in inter-frame positions need to be excluded. For example, data segments with an inter-frame difference less than 2 cm are not included in the sequence construction range. The coordinate data retained after filtering is the effective data of gait changes, forming a dance gait three-dimensional coordinate sequence.

[0061] The spatial vector construction sub-module calculates and constructs two groups of spatial vectors according to the dance gait three-dimensional coordinate sequence, calculates the included angle values within each group of three frames, integrates all the included angle results between three frames, and generates a spatial gait included angle value sequence;

[0062] After obtaining the dance gait three-dimensional coordinate sequence, it is necessary to construct two groups of vectors from three consecutive frames in the sequence. Each group consists of the midpoint coordinates of the front and back two frames to form a spatial vector. For example, in the 28th to 30th frames, take the midpoint coordinates of the 28th and 29th frames to form the first vector , take the 29th and 30th frames to form a second vector , calculate the direction and length of the two vectors respectively. According to the cosine formula of the included angle of three-dimensional space vectors calculate the included angle of the vectors , this included angle is the spatial folding angle formed by the dance movement in this frame paragraph. If indicates that the movement converges. If then it indicates a relatively large turning and opening amplitude. To avoid ambiguous definitions, a boundary reference interval for the included angle value needs to be set, where is the normal swing interval of the dance step. Below this range is the closed posture, and above this interval is the large swing leg movement. The system will perform the above included angle calculation on each group of three frames in turn and generate the corresponding included angle value data to form a spatial gait included angle value sequence. The included angle value calculation involves the inner product operation and modulus length calculation process of vectors. For example:

[0063] ;

[0064] The modulus length is:

[0065] ;

[0066] Performing this type of operation on all frame segment data can obtain a complete spatial gait included angle value sequence.

[0067] The angle mutation judgment sub-module, according to the spatial gait included angle value sequence, uses the formula:

[0068] ;

[0069] Calculate the included angle value of the th frame. According to the frame index where the angle mutation amplitude value is greater than the angle mutation threshold, extract and record the frame position to obtain and establish a turning frame index sequence. Among them, is the spatial vector constructed for the th frame, is the modulus length, is the distance between the two feet key points of the th frame, is the time stamp of the th frame, is the angle mutation amplitude value of the th frame, is the angle mutation amplitude value of the th frame, is the distance between the two feet key points of the th frame, is the time stamp of the th frame;

[0070] According to the spatial gait included angle value sequence, with the included angle value of each frame Calculate the difference between it and the average angle of adjacent frames , and further perform weighted synthesis by combining the inner product of the frame vector, the vector norm, and the distance change trend of the gait key points in this frame to determine whether an angle mutation occurs, where the set threshold = 15° is based on the following: In the continuous stepping movements of ethnic dances, the turning of the dancer's body posture is usually manifested as a significant increase in the spatial angle change. Combining the time span and action rhythm comparison under the condition of a video frame rate of 10fps, in the same dance segment, when the angle change value exceeds 12°, the gait amplitude trend changes, and when the angle change exceeds 15°, it is accompanied by a change in the foot support surface or gait undulation. Therefore, based on this interval, is set to 15°. This value fluctuates with the dancer's step frequency (i.e., the number of steps per unit time) and the movement speed of key points between frames. Its change trend can be adjusted by the median level of the frame segment with the largest angle change per unit time within the dance segment as the dynamic change adjustment benchmark to ensure capturing the key inflection points corresponding to structural action changes. If , then this frame is determined to be a turning frame. The time interval is usually 0.1s (when the frame rate is 10fps). Substitute the data in the table for calculation. The specific calculation of the 29th frame is:

[0071] ;

[0072] Calculate successively to obtain:

[0073] ;

[0074] This value is significantly greater than the set threshold of 15°. Therefore, the 29th frame is marked as a turning frame, and the frame index is automatically recorded to establish a turning frame index sequence.

[0075] Table 1 Angle judgment calculation parameter table:

[0076] ;

[0077] As shown in Table 1, extract the angles, vectors, and spatial parameters of three consecutive frames for subsequent judgment.

[0078] Table 2 Angle mutation amplitude value calculation result table:

[0079] ;

[0080] Table 2 gives the angle mutation amplitude value of the 29th frame. The result shows that this frame is the key position point of the structural change of the dance steps and can be marked as a turning frame and participate in subsequent recognition operations.

[0081] Please refer to Figure 3 , the rhythm segment optimization module includes:

[0082] The beat feature extraction sub-module extracts the time interval between adjacent turning frames based on the turning frame index sequence to obtain the beat duration, and calls the three-dimensional coordinate information within the corresponding frames to calculate the beat amplitude, generating a rhythm segment beat feature sequence;

[0083] Based on the turning frame index sequence, the timestamp information between adjacent frames is sequentially extracted, and the time difference between the two frames is calculated to represent the beat duration. For example, in rhythm segment D1, the turning frame timestamps are 12.0 seconds and 13.5 seconds respectively, and the time interval is 1.5 seconds, indicating that the beat duration of this segment is 1.5 seconds; then, according to the three-dimensional position information of the frames within this turning interval, the maximum and minimum values of the foot z-axis direction coordinates are extracted, and their difference is calculated as the beat amplitude. For example, if the foot is lifted from 0.05 meters to 0.45 meters in the z-axis direction, the beat amplitude of this segment is 0.40 meters. Subsequently, in rhythm segment D2, the timestamp changes from 15.0 seconds to 17.0 seconds, the beat duration is 2.0 seconds, and the corresponding amplitude changes from 0.18 meters to 0.50 meters, with an amplitude of 0.32 meters. In rhythm segment D3, the time span is 1.95 seconds and the amplitude is 0.30 meters. The same operation is performed on all rhythm segments in sequence to generate the beat feature data set shown in Table 3;

[0084] Table 3 Rhythm Segment Parameter Setting Table:

[0085] ;

[0086] As shown in Table 3, each rhythm segment includes the maximum and minimum values of the beat duration and the maximum and minimum values of the beat amplitude. Based on this, the parameter extraction of each segment is completed to generate a rhythm segment beat feature sequence.

[0087] The beat difference calculation sub-module calculates the differences between the maximum and minimum values of the beat duration set and the amplitude set within each segment according to the rhythm segment beat feature sequence, using the formula:

[0088] ;

[0089] Calculate the beat difference combination value of the th segment , to obtain a beat difference combination value sequence, where , respectively represent the maximum and minimum values of the beat duration of the th segment ,

[0090] According to the rhythm section beat feature sequence, calculate the difference between the maximum and minimum values of the duration and amplitude of each beat respectively, and substitute them into the combined calculation formula. For example, for rhythm section D1, given , , , , substituting into the formula gives:

[0091] ;

[0092] ;

[0093] For rhythm section D2, substituting , , , , substituting into the formula gives:

[0094] ;

[0095] ;

[0096] Performing the same calculation on rhythm section D3 gives , and then obtaining the beat difference combination value sequence.

[0097] The stable paragraph screening sub-module determines whether the combined value in each rhythm section is less than the beat stability threshold according to the beat difference combination value sequence, screens all rhythm sections that meet the conditions, and establishes a beat stable paragraph;

[0098] According to the sequence of beat difference combination values, the beat stability threshold is set to 0.7, and each rhythm segment is judged in turn. If its difference combination value is less than the beat stability threshold, it is judged as a stable segment. For example, the combination values of rhythm segments D1, D2, and D3 are 0.3765, 0.4449, and 0.53 respectively, all of which are lower than the threshold and are all marked as stable segments. Finally, their numbers and corresponding parameters are sorted and incorporated into the structure set to obtain the beat stable segments. Among them, the setting basis for the beat stability threshold of 0.7 is the analysis result of the joint distribution interval of the maximum and minimum differences of the beat duration and the amplitude difference in multiple dance sequences. It is extracted by comparison in 90 groups of ethnic dance samples. The statistical mean of the difference combination value is 0.48, and the standard deviation is 0.11. The beat stability threshold is set to the upper limit of the interval of the mean plus 2 times the standard deviation, that is, the beat stability threshold = 0.48 + 2×0.11 = 0.7, to ensure that the selected beat segments have high consistency; this threshold can be adjusted according to the average step size and the action cycle in different types of actions. When the step size is larger and the action cycle is longer, the volatility of the beat amplitude and duration is higher, and the beat stability threshold can be appropriately increased. For example, in the analysis of intense dance segments, the beat stability threshold can be set to 0.8 to 0.85, while in slow rhythm segments, the beat stability threshold should be taken as 0.6 to 0.65, so as to adjust the threshold range according to the rhythm structure of the dance type and ensure that it accurately identifies the structurally stable sections in the action beat distribution.

[0099] Please refer to Figure 4 , the trend stability recognition module includes:

[0100] The coordinate extraction sub-module extracts all frame indexes corresponding to each segment according to the beat stable segments, obtains the three-dimensional spatial position coordinates corresponding to the frame indexes, combines and records the coordinate sequences of each segment in chronological order, and generates the spatial coordinate data of the stable segments;

[0101] According to the set of rhythm-stable paragraphs, first extract all the index numbers from the start frame to the end frame in each paragraph. Combining with the continuous time series information of ethnic dance steps, paragraph 1 can be taken as frame numbers 120 to 140, paragraph 2 as frame numbers 141 to 158, and paragraph 3 as frame numbers 159 to 180. After extracting these frame indices paragraph by paragraph, use a motion capture system or a frame image reconstruction module to obtain the three-dimensional spatial coordinates of the key parts of the dancer's body. The coordinate items are generally the spatial positions in the x, y, and z directions with the unit of meters. For example, the coordinates of the left ankle in the 125th frame can be (0.36, 0.12, 0.08), and the 126th frame is (0.37, 0.12, 0.09). During the coordinate acquisition process, normalize the spatial trajectory changes between adjacent frames so that the position coordinates of each frame within each paragraph correspond one by one to the time series within this paragraph. Then, construct a spatial position matrix in the order of paragraph division. The data within each paragraph consists of time and three-dimensional coordinates. For example, the time series in paragraph 1 is [0, 1, 2,..., 19], and the corresponding coordinate series is , and summarize the coordinate sequences of each paragraph as input data to obtain the spatial coordinate data of the stable paragraphs.

[0102] Based on the spatial coordinate data of the stable paragraphs, the trend fluctuation calculation sub-module performs a linear fitting operation on the sequence of three-dimensional spatial coordinates changing with time in each paragraph to obtain the linear fitting slope value in the main time axis direction of each paragraph. Compare the change intensities between all slope values, and establish a composite calculation relationship for the slope change amount and paragraph length between adjacent paragraphs. Use the formula:

[0103] ;

[0104] Calculate the combined trend fluctuation value , and extract the stable trend segments according to the continuous segment number interval with the smallest fluctuation value to obtain the trend stable interval. Among them, is the linear fitting slope value of the th segment, is the linear fitting slope value of the th segment, is the number of frames of the th segment, is the cumulative operation from the 1st segment to the th segment, and the denominator term is used to avoid division by zero, is the absolute value of the product of the slope value and the inverse segment length, used to measure the sensitive response of trend instability within a short paragraph, represents the total number of segments;

[0105] Based on the stable paragraph spatial coordinate data, perform linear fitting on the sequence of frame coordinates within each paragraph in the main axis direction. The main axis direction is taken as the x-axis of the main movement direction of the dancer, and the fitting slope represents the rate of change of the spatial movement trend in this section. For example, in paragraph 1, the linear fitting slope , in paragraph 2 , in paragraph 3 , in paragraph 4 . Record the fitting slopes of each section and count their corresponding frame lengths L to obtain , , , . As shown in Table 4, further calculate the trend fluctuation combination values between adjacent paragraphs.

[0106] Table 4 Example table of stable paragraph slope values and frame lengths:

[0107] ;

[0108] As shown in Table 4, list the linear fitting slope values of each section in the stable section and their corresponding frame lengths.

[0109] Substitute the above data into the formula for calculation respectively:

[0110] The first term:

[0111] ;

[0112] ;

[0113] ;

[0114] The second term:

[0115] ;

[0116] ;

[0117] ;

[0118] The third term:

[0119] ;

[0120] ;

[0121] ;

[0122] Add up all the values to get the total sum:

[0123] ;

[0124] Therefore, the combined trend fluctuation value of this segment is 0.00253. As shown in Table 5, further sort the values of all segments, and filter the minimum value interval for the next stage of operation to generate a sequence of trend-stable intervals.

[0125] Table 5 Process parameter table for calculating combined trend fluctuation values:

[0126] ;

[0127] Table 5 lists the parameters and intermediate values used in the formula for calculating the combined trend fluctuation value.

[0128] The node index generation sub-module extracts the set of all frame indexes in each stable segment according to the trend-stable interval, calculates the median position according to the time position, extracts the corresponding frame index, and establishes trend-stable node index data;

[0129] According to the sequence of trend-stable intervals, for example, when the combined fluctuation value of paragraph 2 and paragraph 3 is the lowest, extract the set of time positions corresponding to this interval, that is, frame numbers 141 to 180. The total number of frames is 40 frames, so the middle position frame is frame number 141 + (180 - 141) / 2 = 160.5. Extract the frame index and record it as the trend node index value. Traverse all trend-stable intervals in this way to obtain the corresponding middle position frame numbers respectively, and finally establish trend-stable node index data.

[0130] Please refer to Figure 5 , the action information acquisition module includes:

[0131] The main limb coordinate extraction sub-module extracts the set of all frame indexes of the rhythm segment to which the node belongs according to the trend-stable node index data, obtains the three-dimensional position coordinates of the main limb in each frame, and arranges them into a continuous coordinate sequence arranged in frame order to generate a main limb time-series coordinate sequence;

[0132] According to the trend-stationary node index data, extract the rhythm segment numbers corresponding to each paragraph. In each rhythm segment, obtain the three-dimensional coordinate values of the main limb in the order of frames, map and store each frame number with its corresponding coordinates. In a specific implementation, as shown in Table 1, the main limb coordinate data of 5 frames in a certain rhythm segment is recorded. The units of its X, Y, and Z axes are meters. This data is extracted frame by frame through an optical motion capture device and unified into values in an absolute coordinate system. The coordinate points are arranged in chronological order. The first operation is to read the frame numbers corresponding to all node indexes and find the rhythm segment they belong to. Then, filter out all frame sequences of this segment through the paragraph number, such as frames 1 to 5. Then, extract the position values of the main limb in the three-dimensional coordinates for each frame to form a time-continuous coordinate sequence. In this example, the main limb may be the right shoulder joint. After obtaining its three-dimensional space coordinates from frame 1 to frame 5 respectively, the generated time series coordinate sequence of the main limb is a vector group, whose length is the same as the number of frames in the paragraph. Through this sequence, path and direction calculations can be further performed later to obtain the time series coordinate sequence of the main limb.

[0133] Table 6 Example table of spatial coordinates of main limb frame points:

[0134] ;

[0135] As shown in Table 6, this paragraph contains five frames, and the three-dimensional coordinates of the main limb show an upward trend along the time line.

[0136] Based on the time series coordinate sequence of the main limb, calculate the spatial distance between adjacent frame points in sequence according to the frame order, and extract the coordinate difference between the first frame and the last frame to represent the direction. The formula is used:

[0137] ;

[0138] Calculate the path direction combination value , revealing the total amount of action displacement and direction change, and obtaining the spatial motion parameters of the main limb. Among them, is the frame and the frame, the three-dimensional distance between them, , are the coordinate values of the main limb on the x-axis of the first frame and the last frame, is the total number of frames in the rhythm segment;

[0139] Based on the time series coordinate sequence of the main limb, calculate the spatial path distance between adjacent frames, and extract the X coordinate values of the first frame and the last frame to construct the direction term. The path distance calculation uses the three-dimensional Euclidean distance. For frames 1 to 5 in this example, calculate the distances between frame 1 and frame 2, frame 2 and frame 3, frame 3 and frame 4, and frame 4 and frame 5 respectively. Let the distance between each pair of adjacent frames be , and calculate using the formula.

[0140] The path distance is:

[0141] ;

[0142] ;

[0143] ;

[0144] ;

[0145] Then the path accumulation term is: ;

[0146] The direction term is:

[0147] ;

[0148] The final combined value is:

[0149] ;

[0150] This result indicates that there is a combined movement amount of 0.703 meters in the path and direction of the main limb within this paragraph. This value will subsequently be used as a spatial feature of the main limb to participate in the construction of the main limb spatial movement parameters.

[0151] The action information construction sub-module constructs corresponding structured fields and summarizes them based on the main limb spatial movement parameters, combined with the starting frame number and the number of frames in each segment, to establish the main limb action information;

[0152] According to the main limb spatial movement parameters, extract the starting frame index and the total number of frames in each segment, and construct a unified action information data field structure on this basis, which includes path direction combined values, starting frame numbers, number of frames, average displacement, etc. Finally, it is assembled into a single-segment main limb action information object. For example, if the starting frame number of the current segment is 1, the total number of frames is 5, and the path direction combined value is 0.703 meters, it can be represented as a set of parameter data collections. The constructed structure is as follows:

[0153] Starting frame number: 1;

[0154] Number of frames: 5;

[0155] Path direction combined value: 0.703;

[0156] Average path displacement: 0.703 ÷ 5 ≈ 0.1406;

[0157] In this way, the main limb action fields of each rhythm segment are uniformly constructed to generate the main limb action information.

[0158] Please refer to Figure 6 , the synchronization recombination execution module includes:

[0159] The direction correction sub-module traverses all frame points within the current rhythm segment frame by frame according to the main limb motion information, extracts the spatial direction vector of the main limb for each frame, calculates the angular difference between the current direction and the starting direction, and adjusts the current frame direction vector to be consistent with the initial direction, generating a sequence of direction-consistent vectors;

[0160] The operation of traversing all frame points in the current rhythm segment frame by frame according to the main limb motion information needs to be processed in combination with the direction vector information of each frame in the motion frame structure. In a specific implementation, first, the sequence of frame numbers involved in the current rhythm segment is extracted. For example, a frame group numbered from 1 to 5, each frame contains the three-dimensional coordinate points of the main limb. The unit direction vector is calculated by the first frame and the next frame and set as the reference direction. Then, the direction angle is calculated for each subsequent frame in turn. The angle calculation can be performed through the dot product of vectors. For example, in frame 2, the direction of the first frame is set as the vector (1, 0, 0), and the direction of frame 2 is the unit vector (0.996, 0.087, 0). The dot product of the two vectors is 0.996, and the corresponding angle is about 5°. Therefore, the rotation angle difference of frame 2 is 5°. The direction of frame 3 is (0.984, 0.173, 0), the dot product is 0.984, and the angle is about 10°. And so on, the rotation angle of frame 4 is obtained as 7°, and that of frame 5 is 2°. The respective angle values are stored in the rotation angle sequence in turn. Subsequently, the direction correction operation is performed. The current direction vector of each frame is rotationally corrected according to the starting direction angle. The correction method uses a planar rotation operation around the vertical axis, and the coordinate rotation is updated by constructing a rotation transformation matrix. The rotated direction vector of each frame is rewritten into the data structure of that frame, and the direction sequence is updated synchronously, so that all frame direction vectors are aligned to the starting frame direction, obtaining frame vector data with unified directions and generating a sequence of direction-consistent vectors.

[0161] The time scaling sub-module performs an equal-proportion scaling operation on the time duration value corresponding to each frame based on the sequence of direction-consistent vectors, and reallocates the new time point position corresponding to each frame according to the actual time lengths of the starting time of the rhythm segment and the beat period, obtaining an equally aligned time sequence;

[0162] After obtaining the sequence of vectors with the same direction, to unify the time order and rhythm consistency of each frame, a proportional transformation is performed on its original time points. In a specific implementation, first, the original time points are extracted. For example, the sampling times of the current frame 1 to frame 5 are 0ms, 200ms, 400ms, 600ms, and 800ms respectively, and the total time span of the rhythm segment is 800ms. It is set that the rhythm segment should match a periodic beat of 1000ms. Therefore, the 800ms original interval needs to be mapped proportionally to the 1000ms interval. Perform an equal-proportion scaling. Frame 2 is adjusted to (200 / 800)×1000 = 250ms, frame 3 is (400 / 800)×1000 = 500ms, frame 4 is 750ms, and frame 5 is 1000ms. The scaled time points are rewritten into the corresponding data fields of each frame. Combining parameters such as the frame number, starting direction angle, current direction angle, and rotation angle difference, a synchronous timing structure is formed. The specific data is shown in the following table:

[0163] Table 7 Example table of frame-level direction and time rearrangement of ethnic dance:

[0164] ;

[0165] As shown in Table 7, after completing the scaling process of the time points of each frame, the updated time points can be strictly aligned with the rhythm segment periodic line, and finally an equal-proportion aligned time sequence is obtained.

[0166] The structure rearrangement sub-module rearranges the order of all frames within the rhythm segment according to the equal-proportion aligned time sequence, and at the same time reconstructs the coordinates, directions, and time tags corresponding to each frame, and organizes them into a continuous dynamic structure data uniformly to obtain the dynamic display result of ethnic dance;

[0167] According to the equal-proportion aligned time sequence, the frame order and each parameter field are reorganized and constructed. First, sort the frame numbers in ascending order of the scaled time points. The time of frame 1 is 0ms, the time of frame 2 is 250ms, and so on to 1000ms of frame 5. The corresponding frame number order does not need to be adjusted, but the coordinate and direction data need to be repackaged. Combine the three-dimensional space coordinates, vectors with the same direction, and scaled time points of each frame after sorting into a unified frame data structure. Each frame structure contains fields such as frame number, three-dimensional coordinate points, direction vectors, and scaled time stamps. This data structure supports sequential calls during subsequent dynamic display rendering. In addition, form a time-ordered continuous dynamic set of all frame data, and finally it can be output to a three-dimensional action display module or a rendering engine for display to establish the dynamic display result of ethnic dance.

[0168] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the relevant art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A digital display system for ethnic dances, characterized in that The system includes: The trajectory node extraction module obtains the three-dimensional position coordinates of ethnic dance movements, constructs spatial vectors, calculates the included angle values, judges and records the angle mutation points, and generates a turning frame index sequence; The rhythm segment optimization module, based on the turning frame index sequence, obtains the beat duration and beat amplitude of the turning paragraphs, calculates the difference between the maximum and minimum values of the beat duration and amplitude respectively, compares with the beat stability threshold, and filters out the beat-stable paragraphs; The trend stability recognition module, according to the beat-stable paragraphs, extracts the spatial position coordinates of the corresponding frame points, performs linear fitting and extracts the corresponding fitting slopes, judges the continuous paragraphs with the smallest fluctuation amplitude and extracts the frame indexes, and generates the trend-stable node index data; The motion information acquisition module, according to the trend-stable node index data, obtains the three-dimensional position points of each frame of the main limb, calculates the total path distance, constructs a direction vector and calculates the direction angle, and combines the starting frame and the total number of frames to obtain the main limb motion information; The synchronous recombination execution module, according to the main limb motion information, calculates the rotation angle and performs direction correction, scales the time duration values of all frames proportionally, rearranges the frame sequence to construct a new dynamic structure, and obtains the ethnic dance dynamic display result; The trend stability recognition module includes: The coordinate extraction sub-module, according to the beat-stable paragraphs, extracts all the frame indexes corresponding to each segment, obtains the three-dimensional spatial position coordinates corresponding to the frame indexes, combines and records the coordinate sequences of each segment in chronological order, and generates the stable paragraph spatial coordinate data; The trend fluctuation calculation sub-module, based on the stable paragraph spatial coordinate data, performs a linear fitting operation on the sequence of three-dimensional spatial coordinates changing with time in each paragraph, obtains the linear fitting slope value in the main time axis direction of each paragraph, compares the change intensity between all slope values, and establishes a composite calculation relationship for the slope change amount and paragraph length between adjacent paragraphs, using the formula: ； Calculate the combined trend fluctuation value , extract the stable trend segment according to the continuous segment number interval with the smallest fluctuation value to obtain the trend stable interval, where is the linear fitting slope value of the th segment, is the linear fitting slope value of the th segment, is the number of frames of the th segment, is the cumulative operation from the 1st segment to the th segment, is the absolute value of the product of the slope value and the inverse segment length, and n represents the total number of segments; The node index generation sub-module, according to the trend-stable interval, extracts the set of all frame indexes in each stable segment, calculates the median position according to the time position, extracts the corresponding frame indexes, and establishes the trend-stable node index data.

2. The digital display system for ethnic dances according to claim 1, characterized in that, The turning frame index sequence includes action change nodes, turning angle distributions, and action segmentation boundaries. The beat-stable paragraphs include beat balance indicators, rhythm start and end information, and stability interval identifiers. The trend-stable node index data includes stable trend reference frames, fitting change ranges, and rhythm trend parameters. The main limb motion information includes action path characteristics, direction change parameters, and action timing structures. The ethnic dance dynamic display result includes a recombined action frame sequence, a unified direction and posture, and a beat-synchronized rhythm.

3. The digital display system for ethnic dance according to claim 1, characterized in that The trajectory node extraction module includes: The three-dimensional coordinate extraction sub-module, based on the video frame sequence of continuous stepping movements of ethnic dance, obtains the three-dimensional position coordinate data of the dancer's bones in each frame, sequentially extracts the three-dimensional coordinates of the dancer's feet joint points in each frame, organizes and stores them as a three-dimensional coordinate set sorted by frame order, and generates a dance gait three-dimensional coordinate sequence; The spatial vector construction sub-module constructs two sets of spatial vectors based on the three-dimensional coordinate sequence of the dance gait, calculates the included angle values within each set of three frames, integrates the included angle results between all three frames, and generates a spatial gait included angle value sequence; The angle mutation judgment sub-module is based on the spatial gait included angle value sequence and uses the formula: ； Calculate the included angle value of the frame. According to the frame index where the angle mutation amplitude value is greater than the angle mutation threshold, extract and record the frame position, obtain and establish a sequence of turning frame indexes. Among them, is the spatial vector constructed for the frame, is the modulus length, is the distance between the two feet key points of the frame, is the time stamp of the frame, is the angle mutation amplitude value of the frame, is the angle mutation amplitude value of the frame, is the distance between the two feet key points of the frame, is the time stamp of the frame.

4. The digital display system for ethnic dances according to claim 1, characterized in that, The rhythm segment optimization module includes: The beat feature extraction sub-module extracts the time interval between adjacent turning frames based on the turning frame index sequence to obtain the beat duration, calls the three-dimensional coordinate information within the corresponding frames to calculate the beat amplitude, and generates a rhythm segment beat feature sequence; The beat difference calculation sub-module calculates the differences between the maximum and minimum values of the beat duration set and the amplitude set within each segment based on the rhythm segment beat feature sequence, and uses the formula: ； Calculate the combined value of the beat difference for the i-th segment , to obtain a sequence of combined values of beat differences, where , respectively represent the maximum and minimum values of the beat duration for the i-th segment, , respectively represent the maximum and minimum values of the beat amplitude for the i-th segment; The stable paragraph screening sub-module determines whether the combined value in each rhythm segment is less than the beat stability threshold based on the beat difference combined value sequence, screens all rhythm segments that meet the conditions, and establishes a beat stable paragraph.

5. The digital display system for ethnic dances according to claim 1, wherein The action information acquisition module includes: The main limb coordinate extraction sub-module extracts all frame indexes of the rhythm segment to which the node belongs based on the trend stable node index data, obtains the three-dimensional position coordinates of the main limb in each frame, and arranges them into a continuous coordinate sequence in frame order, generating a main limb time series coordinate sequence; The action path calculation sub-module calculates the spatial distance between adjacent frame points in sequence according to the frame order based on the main limb time series coordinate sequence, extracts the coordinate difference between the first frame and the last frame to represent the direction, and uses the formula: ； Calculate the combined value of the path direction , reveal the total amount of movement displacement and direction changes, and obtain the main limb spatial motion parameters, where is the three-dimensional distance between the th frame and the , are the coordinate values of the main limb on the x-axis of the first frame and the last frame, and m is the total number of frames in the rhythm segment; The action information construction sub-module constructs corresponding structured fields and summarizes them based on the main limb spatial motion parameters in combination with the starting frame number and the number of frames in each segment, and establishes the main limb action information.

6. The digital display system for ethnic dances according to claim 1, characterized in that The synchronous recombination execution module includes: The direction correction sub-module traverses all frame points within the current rhythm segment one by one based on the main limb action information, extracts the spatial direction vector of the main limb in each frame, calculates the included angle difference between the current direction and the starting direction, and adjusts the current frame direction vector to be consistent with the initial direction, generating a direction consistent vector sequence; The time scaling sub-module performs an equal ratio scaling operation on the time duration value corresponding to each frame based on the direction consistent vector sequence, and reallocates the new time point position corresponding to each frame according to the actual time lengths of the rhythm segment start time and the beat period, obtaining an equal ratio alignment time sequence; The structure rearrangement sub-module rearranges the order of all frames within the rhythm segment according to the equal ratio alignment time sequence, and simultaneously reconstructs the coordinates, directions, and time tags corresponding to each frame, and organizes them into a continuous dynamic structure data uniformly to obtain the ethnic dance dynamic display result.

Images