Method and equipment for realizing intelligent stage performance and medium

Abstract

The invention relates to the field of stage lighting, and particularly discloses a method and device for achieving intelligent stage performance and a medium, and the method comprises the steps: building a pose prediction model for predicting the pose data of a performer after a first time length through a target detection algorithm, a target tracking algorithm, a human body pose estimation algorithm and a time sequence prediction model; by utilizing a machine learning algorithm and an illumination parameter formula, establishing and training to obtain a lighting effect model for calculating stage lamp control data through the pose data of the performer; according to the real-time pose data, obtained by the sensor nodes, of the performer, stage lamp prediction control data are obtained through calculation, and therefore stage lamps are controlled. According to the invention, intelligent control of stage lamps is realized, real-time monitoring and tracking of performers are realized, and light and shadow can be ensured to accurately follow the action change of the performers by predicting the trajectories of the performers in advance.

Description

Technical Field

[0001] The present invention relates to the field of stage lighting, and in particular to a method, device and medium for realizing intelligent stage performance. Background Art

[0002] With the rapid development of modern stage equipment and the rise of artificial intelligence technology, a foundation has been laid for the development of intelligent stage performance technology. Intelligent stage performance is an innovative solution that combines the intelligent tracking method of spotlights with stage performances. By applying intelligent algorithms and sensor technology to the stage lighting system, performers can be tracked more accurately, thereby improving the performance effect and the audience's visual experience, and avoiding the disconnection between the preset lighting control and the performance caused by the performers' improvisation or mistakes.

[0003] However, existing technical solutions or public materials related to intelligent stage performance technology, such as the Chinese invention patent (application number 202410185349.0, a method for intelligent control of stage lighting), also provide technical inspiration for realizing the prediction of performer movements and automatic tracking of light and shadow effects. Through intelligent algorithms, the wonderful areas on the stage are identified and predicted, and the performer's trajectory is predicted, thereby realizing intelligent control of lighting. However, there are various difficult problems in its public solutions, such as a single means of obtaining posture information, a large model size, many variables involved, and a wide range of coverage. The model predicts wonderful areas with at least four degrees of freedom, and the correlation between the degrees of freedom is low, resulting in poor learning efficiency and the need for a large number of samples, resulting in uneven prediction results, long calculation cycles, and difficulty in training.

[0004] Moreover, similar solutions are prone to problems such as failure to accurately obtain stage images and thus malfunction when the stage effects are mixed with a large amount of smoke, flames and strobe lights. Moreover, most of the existing similar solutions are still in the conceptual stage and do not have the practicality and conditions for actual implementation. Summary of the invention

[0005] In order to overcome the above problems, the present invention provides a method, device and medium for realizing intelligent stage performance.

[0006] The present invention provides a method for realizing intelligent stage performance, comprising:

[0007] Establishing a posture prediction model for predicting the performer's posture data after a first time length, and obtaining several sets of previous continuous performer's historical posture data as a training set and a verification set to complete the training of the posture prediction model; wherein the posture prediction model is provided with a target detection algorithm, a target tracking algorithm, a human posture estimation algorithm and a time series prediction model, and the first time length is a preset time length;

[0008] Using machine learning algorithms and lighting parameter formulas, a lighting effect model is established for calculating stage light control data through performer posture data; obtaining several sets of stage light control data and their corresponding performer posture data as training sets and verification sets to complete the training of the lighting effect model;

[0009] A plurality of sensor nodes are arranged around the stage, and the position information and motion data of the performers are continuously acquired by using the plurality of sensor nodes; wherein the sensor nodes include one or any combination of a camera, a depth sensor, a motion capture sensor and an infrared sensor;

[0010] According to the installation positions of several sensor nodes, the position information and motion data are preprocessed to obtain the performer's real-time posture data;

[0011] Importing the real-time posture data into the posture prediction model for calculation to obtain predicted posture data, and importing the predicted posture data into the lighting effect model to obtain stage light prediction control data;

[0012] The stage light prediction control data is used as the stage light control data after the first time length in the future to control the stage light.

[0013] Preferably, the data preprocessing includes filtering of position information, and denoising, edge detection and color space conversion of motion data.

[0014] Preferably, the use of a target detection algorithm, a target tracking algorithm, a human posture estimation algorithm and a time series prediction model to establish a posture prediction model for predicting the performer's posture data after the first time length is specifically as follows:

[0015] Use the target detection algorithm YOLO to identify the performer’s human target on the stage and output it as a target box;

[0016] Use the target tracking algorithm MOSSE to track the trajectory data of the target box;

[0017] Using the human pose estimation algorithm OpenPose, the coordinates of the performer's skeleton key points are identified from the target frame and output as human pose data;

[0018] Using the human body posture data and trajectory data as the position and posture data of the performer;

[0019] According to the performer's continuous posture data, the time series prediction model LSTM is used to predict the target's posture data after the first time length in the future.

[0020] Preferably, the illumination parameter formula is specifically:

[0021] P=wA ·A+w L ·L+w B ·B+w S ·S

[0022] Among them, P represents the lighting parameter, A represents the influence coefficient of the action type, and W A represents the action weight, L represents the influence coefficient of the performer's position, and W L represents the position weight, B represents the background influence coefficient, W B represents the background weight, S represents the influence coefficient of performance style, and W S Represents style weight.

[0023] Preferably, the method further comprises the following steps:

[0024] Several preset strategies are set, and the preset strategies are used to enable the model to limit the stage light control data of specific performer posture data before model training.

[0025] Preferably, the method further comprises the following steps:

[0026] Record all the performer posture data and stage light control data in the current performance, and output them as a command timeline;

[0027] The stage light control data corresponding to the posture data of several performers in the timeline are modified and output as manual control instructions, and the execution priority of the manual control instructions is made higher than that of the stage light prediction control data.

[0028] Preferably, the method further comprises the following steps:

[0029] Identify the performer’s identity information through facial recognition algorithms;

[0030] Different pose prediction models and lighting effect models are trained and executed for different performers.

[0031] The present invention also provides a device for realizing intelligent stage performance, comprising: a posture prediction module, a lighting effect module, an acquisition module, a posture calculation module, a lighting effect calculation module and a lighting control module;

[0032] The posture prediction module is used to establish a posture prediction model for predicting the performer's posture data after a first time length, and obtain several previous continuous performer's historical posture data as a training set and a verification set to complete the training of the posture prediction model; wherein the posture prediction model is provided with a target detection algorithm, a target tracking algorithm, a human posture estimation algorithm and a time series prediction model, and the first time length is a preset time length;

[0033] The lighting effect is used to establish a lighting effect model for calculating stage light control data through performer posture data by using a machine learning algorithm and a lighting parameter formula; a plurality of previous stage light control data and their corresponding performer posture data are obtained as a training set and a verification set to complete the training of the lighting effect model;

[0034] The acquisition module is used to set up a number of sensor nodes around the stage, and use the several sensor nodes to continuously acquire the position information and motion data of the performers; wherein the sensor nodes include one of a camera, a depth sensor, a motion capture sensor and an infrared sensor or any combination thereof;

[0035] The posture calculation module is used to perform data preprocessing on the position information and motion data according to the installation positions of a plurality of sensor nodes, so as to obtain the real-time posture data of the performer;

[0036] The lighting effect calculation module is used to import the real-time posture data into the posture prediction model for calculation to obtain predicted posture data, and import the predicted posture data into the lighting effect model to obtain stage light prediction control data;

[0037] The lighting control module is used to use the stage light prediction control data as the stage light control data after a first time length in the future to control the stage light.

[0038] The present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned method for realizing intelligent stage performance. The beneficial effects of the present invention are:

[0039] (1) Through the pose prediction model formed by combining the target detection algorithm, target tracking algorithm, human pose estimation algorithm and time series prediction model, real-time monitoring and tracking of performers can be achieved. By predicting the performer's trajectory in advance, it ensures that the light and shadow can accurately follow the performer's movements. The performer's pose is obtained through a variety of sensor parameters. It can be widely used in various performance occasions such as stage plays, concerts, dance performances, etc., bringing new possibilities and innovations to stage art.

[0040] (2) Through the combination of machine learning algorithms and lighting parameter formulas, intelligent control of stage lights is achieved, which saves labor costs, improves the work efficiency of lighting engineers, enhances the audience's sense of participation and immersion, and improves the overall viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention will be further described below in conjunction with the accompanying drawings, wherein:

[0042] Figure 1 This is a flow chart of a method according to Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 As a first embodiment of the present invention, a method for realizing intelligent stage performance is disclosed, comprising:

[0045] S1, establishing a posture prediction model for predicting the posture data of the performer after a first time length, and obtaining several previous continuous historical posture data of the performer as a training set and a verification set to complete the training of the posture prediction model; wherein the posture prediction model is provided with a target detection algorithm, a target tracking algorithm, a human posture estimation algorithm and a time series prediction model, and the first time length is a preset time length;

[0046] S2, lighting effects, using machine learning algorithms and lighting parameter formulas, and obtaining several sets of stage light control data and their corresponding performer posture data as training sets and validation sets, to establish and train a lighting effect model for calculating stage light control data through performer posture data; wherein the stage light control data includes the direction, rotation speed, intensity, color and lighting effect of each stage light;

[0047] S3, acquiring data, using a number of sensor nodes around the stage to continuously acquire the position information and motion data of the performers; wherein the sensor nodes include one or any combination of a camera, a depth sensor, a motion capture sensor and an infrared sensor;

[0048] S4, posture calculation, based on the installation positions of several sensor nodes, the position information and motion data are preprocessed to obtain the performer's real-time posture data;

[0049] S5, lighting effect calculation, importing the real-time posture data into the posture prediction model for calculation to obtain predicted posture data, and importing the predicted posture data into the lighting effect model to obtain stage light prediction control data;

[0050] S6. Lighting control: using the stage light prediction control data as the stage light control data after the first time length in the future to control the stage light.

[0051] The data preprocessing includes filtering of position information, and denoising, edge detection and color space conversion of motion data.

[0052] This embodiment also discloses a device for realizing intelligent stage performance, including: a posture prediction module, a lighting effect module, an acquisition module, a posture calculation module, a lighting effect calculation module and a lighting control module;

[0053] The posture prediction module is used to establish a posture prediction model for predicting the posture data of the performer after a first time length by using a target detection algorithm, a target tracking algorithm, a human posture estimation algorithm and a time series prediction model, and obtain several previous continuous historical posture data of the performer as a training set and a verification set to complete the training of the posture prediction model; wherein the first time length is a preset time length;

[0054] The lighting effect is used to utilize a machine learning algorithm and a lighting parameter formula, and obtain several sets of stage light control data and their corresponding performer posture data as a training set and a validation set, to establish and train a lighting effect model for calculating the stage light control data through the performer's posture data;

[0055] The acquisition module is used to continuously acquire the position information and motion data of the performer using a number of sensor nodes around the stage; wherein the sensor nodes include one of a camera, a depth sensor, a motion capture sensor and an infrared sensor or any combination thereof;

[0056] The posture calculation module is used to perform data preprocessing on the position information and motion data according to the installation positions of a plurality of sensor nodes, so as to obtain the real-time posture data of the performer;

[0057] The lighting effect calculation module is used to import the real-time posture data into the posture prediction model for calculation to obtain predicted posture data, and import the predicted posture data into the lighting effect model to obtain stage light prediction control data;

[0058] The lighting control module is used to use the stage light prediction control data as the stage light control data after a first time length in the future to control the stage light.

[0059] This embodiment uses target detection algorithms, target tracking algorithms, human posture estimation algorithms and time series prediction models to achieve real-time monitoring and tracking of performers. It predicts the performer's trajectory in advance to ensure that light and shadow can accurately follow the performer's movements. It obtains the performer's posture through a variety of sensor parameters. It can be widely used in various performance occasions such as stage plays, concerts, and dance performances, bringing new possibilities and innovations to stage art. Through machine learning algorithms and lighting parameter formulas, intelligent control of stage lights is achieved, saving labor costs, improving the work efficiency of lighting technicians, enhancing the audience's sense of participation and immersion, and improving the overall viewing experience.

[0060] The following is Example 2 of the present invention. This embodiment is a more specific application example of Example 1. Through a camera and an infrared sensor, infrared images of the stage screen and the performers are collected. The software architecture adopted is: using Python programming language to build algorithm logic; using OpenCV library for image processing; using TensorFlow, PyTorch and other frameworks to build machine learning models.

[0061] In step S1 of this embodiment, the target detection algorithm, target tracking algorithm, human posture estimation algorithm and time series prediction model are used to establish a posture prediction model for predicting the performer's posture data after the first time length, specifically:

[0062] Use target detection algorithms (such as YOLO, FasterR-CNN, etc.) to identify the performer’s human targets on the stage and output them as target boxes;

[0063] Use target tracking algorithms (such as MOSSE, KCF, etc.) to track the trajectory data of the target box;

[0064] Using human pose estimation algorithms (such as OpenPose, AlphaPose, etc.), the coordinates of the performer's skeleton key points are identified from the target frame and output as human pose data;

[0065] Using the human body posture data and trajectory data as the position and posture data of the performer;

[0066] Based on the performer's continuous posture data, a time series prediction model (such as LSTM, GRU, etc.) is used to predict the posture data of the target after a first time length in the future.

[0067] The motion capture sensor in this embodiment can be a WIFI router that transmits data to each other, and the WiFi.based human motion recognition technology is used to identify the actor's movements on the stage through the dual-stream convolution-enhanced Transformer. Different human body movements will produce "subtle" but "unique" signal changes in the channel state information (CSI) between the signal transmitter (TX) and the receiver (RX); the dual-stream convolution-enhanced Transformer (THAT) model is used to distinguish the original CSI input into a channel stream and a timing stream, and the channel stream and the time stream are processed separately by two parallel neural modules to extract signal change features, thereby performing table verification of position and motion information. The motion capture sensor relies on occlusion during radio transmission to judge the actor's movements and positions, and is not affected by visual occlusions and light changes caused by smoke, lights, etc., so it can adapt to more complex special effects stages.

[0068] The target tracking algorithm adopts the MOSSE algorithm, which is the first to apply correlation filtering to the field of target tracking. While maintaining a good tracking effect, its speed far exceeds other tracking algorithms of the same period, and can reach 669FPS at the fastest.

[0069] Assuming that this embodiment specifically adopts a human posture estimation model based on a convolutional neural network (CNN) and combines a long short-term memory network (LSTM) for action prediction, the process is summarized as follows:

[0070] 1. Human body posture estimation:

[0071] 1.1. Use the OpenPose model to identify the key point coordinates of the performer;

[0072] 1.2. Input video frames to the CNN model and output key point coordinates;

[0073] 2. Action Prediction:

[0074] 2.1. Input the continuous key point coordinate sequence into the LSTM model;

[0075] 2.2. The LSTM model outputs the position and movement trend of the performer in the next few seconds;

[0076] 3. Follow spot control:

[0077] 3.1. Calculate the moving angle of the chase light based on the predicted position information;

[0078] 3.2. Send instructions to the follow spot controller to adjust the direction and height of the follow spot;

[0079] 4. Light and shadow effect adjustment:

[0080] 4.1. Calculate the required light and shadow effect parameters according to the action type and stage background;

[0081] 4.2. Change the color and intensity of the follow spot through API calls or predefined programs.

[0082] With the iteration and innovation of the algorithm, the algorithm in this solution can be replaced by a newer similar algorithm to further improve the prediction accuracy and computational efficiency. Those skilled in the art can understand that all algorithm alternatives based on the same principle are within the scope of protection of this application.

[0083] The illumination parameter formula of this embodiment is specifically:

[0084] P=w A ·A+w L ·L+w B ·B+w S ·S

[0085] Among them, P represents lighting parameters (including intensity, angle, color, etc.), A represents the influence coefficient of action type, WA represents action weight, L represents the influence coefficient of performer position, WL represents position weight, B represents background influence coefficient, WB represents background weight, S represents the influence coefficient of performance style, and WS represents style weight.

[0086] The above formula is a simple linear model, which is used to simplify the model learning and training process of various parameters and lighting effects in the stage. In actual application, the formula can add more complex functions and parameters to enhance the calculation accuracy of lighting parameters.

[0087] The specific implementation goal of the stage light control data in this embodiment is to control the follow-up light and adjust the light and shadow effects:

[0088] 1. Follow spot control: Use the predicted position information to adjust the direction and height of the follow spot in real time to ensure that the light is always focused on the performer; adjust the speed of the follow spot according to the performer's movement speed to maintain stable tracking;

[0089] 2. Lighting and shadow effect adjustment: Combined with the results of action prediction and lighting demand assessment, the lighting angle, intensity and color are automatically adjusted; through preset lighting modes or real-time calculated lighting parameters, unique lighting and shadow effects are created for the performance.

[0090] The method of intelligent stage performance in this embodiment may also include the following steps:

[0091] S7. Setting a number of preset strategies, wherein the preset strategies are used to enable the model to limit the stage light control data of specific performer posture data before model training.

[0092] This step is used to limit the lighting effect changes of the model in a specific scene and improve the accuracy of the model output results. It is convenient for lighting engineers to preset the lighting effects corresponding to some scenes, thereby improving the accuracy of stage light control data prediction.

[0093] The method of intelligent stage performance in this embodiment may also include the following steps:

[0094] S8, recording all the performers' posture data and stage light control data in the current performance, and outputting them as a command timeline;

[0095] S9, modifying the stage light control data corresponding to the posture data of several performers in the timeline, outputting them as manual control instructions, and making the execution priority of the manual control instructions higher than the stage light prediction control data.

[0096] It is convenient for the lighting engineer to observe the automatic control effect of the lighting during the subsequent performance after rehearsal and preview, and insert manual control instructions or modify algorithm parameters in certain time periods to manually optimize the stage effect.

[0097] The method of intelligent stage performance in this embodiment includes the following steps:

[0098] S10, identifying the identity information of the performer through a face recognition algorithm;

[0099] S11. For different performers, train and execute different posture prediction models and lighting effect models respectively.

[0100] In order to arrange different lighting effects for different actors.

[0101] In other embodiments, a closed-loop control system can be established through a feedback mechanism, and the algorithm parameters can be adjusted based on the actual stage lighting effects observed by the lighting engineer or the evaluation model. By continuously optimizing the algorithm performance, the tracking accuracy and stability of the overall model can be improved.

[0102] The following is a third embodiment of the present invention. This embodiment is a more specific application example of the first embodiment. The data set of this embodiment, that is, the training set and the verification set are constructed as follows:

[0103] Data source: Collect a large number of performer action videos, which should cover a variety of dance styles, action types, and performance situations to ensure the generalization ability of the model.

[0104] Data annotation: Use a human pose estimation algorithm (such as OpenPose) to process the video frames and extract the coordinates of the performer's key points. These coordinates should include timestamps to maintain the continuity of the action.

[0105] Sequence data creation: Take a continuous sequence of key point coordinates as the input of the model. The sequence length needs to be determined based on the complexity of the action and the scope of the prediction. For example, use the key point coordinates of the previous 5 seconds to predict the action of the next 1 second.

[0106] Data partitioning: Divide the dataset into training set, validation set and test set, usually in a ratio of 70%, 15%, and 15%. The training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the final performance of the model.

[0107] The return function and standard of the model training in this embodiment are as follows:

[0108] Loss function: LSTM model training usually uses mean squared error (MSE) as the loss function, which measures the difference between the predicted keypoint coordinates and the actual coordinates. For each time step, the loss function can be expressed as:

[0109]

[0110] Among them, yi is the actual key point coordinate, y ^ i is the keypoint coordinate predicted by the model and N is the number of keypoints.

[0111] Regularization term: To avoid overfitting, a regularization term, such as L1 or L2 regularization, can be added to the loss function to penalize large weight values.

[0112] Optimizer selection: Use a gradient descent optimizer, such as Adam or RMSprop, to minimize the loss function. These optimizers can automatically adjust the learning rate to speed up convergence.

[0113] The basic unit of the LSTM model contains a memory unit and three gates (input gate, forget gate, output gate). Each gate is a linear layer plus an activation function (such as sigmoid or tanh) to control the flow of information. The mathematical formula of the LSTM unit is as follows:

[0114] Forget Gate:

[0115] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0116] Where Wf is the weight matrix, bf is the bias vector, ht-1 is the hidden state of the previous time step, xt is the input of the current time step, and σ is the sigmoid activation function.

[0117] Input Gate:

[0118] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0119] Candidate memory cell status:

[0120]

[0121] Memory unit status:

[0122]

[0123] Here, ⊙ represents element-wise multiplication.

[0124] Output Gate:

[0125] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0126] Hidden state:

[0127] h t =o t ☉tanh(c t )

[0128] The model training process is as follows:

[0129] Initialization: Randomly initialize the weights and biases of the model.

[0130] Forward propagation: Input the sequence data into the model and calculate the output using the above formula.

[0131] Backpropagation: Calculate the gradient of the loss function and update the weights.

[0132] Iterative training: Repeat the forward propagation and backpropagation process until the model converges or reaches a predetermined number of training rounds.

[0133] Through the above steps, the LSTM model learns the dynamic patterns of the performer's movements and predicts the position and movement trends in the next few seconds.

[0134] The present invention also discloses a terminal device, including a processor and a storage device, the storage device is used to store one or more programs; when the one or more programs are executed by the processor, the processor implements the above-mentioned method for realizing intelligent stage performance. The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The processor is the control center of the test equipment, and various interfaces and lines are used to connect the various parts of the entire test equipment.

[0135] The storage device can be used to store computer programs and / or modules. The processor realizes various functions of the terminal device by running or executing the computer programs and / or modules stored in the storage device and calling the data stored in the storage device. The storage device can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the terminal device, etc. In addition, the storage device can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0136] Among them, if the module / unit of the device integration for realizing the intelligent stage performance is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in at least one computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0137] It should be noted that the embodiments of the devices and apparatuses described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Claims

1. A method for realizing intelligent stage performance, characterized in that: include: Establishing a posture prediction model for predicting the performer's posture data after a first time length, and obtaining several sets of previous continuous performer's historical posture data as a training set and a verification set to complete the training of the posture prediction model; wherein the posture prediction model is provided with a target detection algorithm, a target tracking algorithm, a human posture estimation algorithm and a time series prediction model, and the first time length is a preset time length; Using machine learning algorithms and lighting parameter formulas, a lighting effect model is established for calculating stage light control data through performer posture data; obtaining several sets of stage light control data and their corresponding performer posture data as training sets and verification sets to complete the training of the lighting effect model; A plurality of sensor nodes are arranged around the stage, and the position information and motion data of the performers are continuously acquired by using the plurality of sensor nodes; wherein the sensor nodes include one or any combination of a camera, a depth sensor, a motion capture sensor and an infrared sensor; According to the relative positions of several sensor nodes, the position information and motion data are preprocessed to obtain the performer's real-time posture data; Importing the real-time posture data into the posture prediction model to obtain predicted posture data, and importing the predicted posture data into the lighting effect model to obtain stage light prediction control data; The stage light prediction control data is used as the stage light control data after the first time length in the future to control the stage light.

2. A method for realizing intelligent stage performance according to claim 1, characterized in that: The data preprocessing includes filtering of position information, and denoising, edge detection and color space conversion of motion data.

3. A method for realizing intelligent stage performance according to claim 1, characterized in that: The target detection algorithm, the target tracking algorithm, the human posture estimation algorithm and the time series prediction model are used to establish a posture prediction model for predicting the performer's posture data after the first time length, specifically: Use the target detection algorithm YOLO to identify the performer’s human target on the stage and output it as a target box; Use the target tracking algorithm MOSSE to track the trajectory data of the target box; Using the human pose estimation algorithm OpenPose, the coordinates of the performer's skeleton key points are identified from the target frame and output as human pose data; Using the human body posture data and trajectory data as the position and posture data of the performer; According to the performer's continuous posture data, the time series prediction model LSTM is used to predict the target's posture data after the first time length in the future.

4. A method for realizing intelligent stage performance according to claim 1, characterized in that: The illumination parameter formula is specifically: P=ω A ·A+ω L ·L+ω B ·B+ω S ·S Among them, P represents the lighting parameters, A represents the influence coefficient of the action type, WA represents the action weight, L represents the influence coefficient of the performer's position, WL represents the position weight, B represents the background influence coefficient, WB represents the background weight, S represents the influence coefficient of the performance style, and WS represents the style weight.

5. A method for realizing intelligent stage performance according to claim 1, characterized in that: The following steps are also included: Several preset strategies are set, and the preset strategies are used to enable the model to limit the stage light control data of specific performer posture data before model training.

6. A method for realizing intelligent stage performance according to claim 1, characterized in that: The following steps are also included: Record all the performer posture data and stage light control data in the current performance, and output them as a command timeline; The stage light control data corresponding to the posture data of several performers in the timeline are modified and output as manual control instructions, and the execution priority of the manual control instructions is made higher than that of the stage light prediction control data.

7. A method for realizing intelligent stage performance according to claim 1, characterized in that: The following steps are also included: Identify the performer’s identity information through facial recognition algorithms; Different pose prediction models and lighting effect models are trained and executed for different performers.

8. A device for realizing intelligent stage performance, characterized in that: include: Posture prediction module, lighting effect module, acquisition module, posture calculation module, lighting effect calculation module and lighting control module; The posture prediction module is used to establish a posture prediction model for predicting the performer's posture data after a first time length, and obtain several previous continuous performer's historical posture data as a training set and a verification set to complete the training of the posture prediction model; wherein the posture prediction model is provided with a target detection algorithm, a target tracking algorithm, a human posture estimation algorithm and a time series prediction model, and the first time length is a preset time length; The lighting effect is used to establish a lighting effect model for calculating stage light control data through performer posture data by using a machine learning algorithm and a lighting parameter formula; a plurality of previous stage light control data and their corresponding performer posture data are obtained as a training set and a verification set to complete the training of the lighting effect model; The acquisition module is used to set up a number of sensor nodes around the stage, and use the several sensor nodes to continuously acquire the position information and motion data of the performers; wherein the sensor nodes include one of a camera, a depth sensor, a motion capture sensor and an infrared sensor or any combination thereof; The posture calculation module is used to perform data preprocessing on the position information and motion data according to the installation positions of a plurality of sensor nodes, so as to obtain the real-time posture data of the performer; The lighting effect calculation module is used to import the real-time posture data into the posture prediction model for calculation to obtain predicted posture data, and import the predicted posture data into the lighting effect model to obtain stage light prediction control data; The lighting control module is used to use the stage light prediction control data as the stage light control data after a first time length in the future to control the stage light.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for realizing intelligent stage performance according to any one of claims 1 to 7.

Images