Naked eye 3D interactive display processing method applied to giant spherical building

Abstract

The invention provides a naked-eye 3D interactive display processing method applied to a giant spherical building, and relates to the technical field of naked-eye 3D display, and the method comprises the following steps: recognizing a display screen group in the giant spherical building, and collecting spatial information to generate a spatial three-dimensional model; generating naked-eye 3D content based on the spatial three-dimensional model, and mapping the naked-eye 3D content on a display screen group; identifying a selected interactive user; establishing an interaction mechanism including a hand recognition mechanism and a voice recognition mechanism; at least one interaction mechanism is selected, interaction information corresponding to the interaction mechanism is recognized and analyzed, and a gesture execution instruction or a response voice instruction is obtained. By establishing an interaction mechanism, through a hand recognition and voice recognition mechanism, a user can naturally interact with the naked-eye 3D content, and by combining an ambient light adaptation mechanism, the change of ambient light is sensed, and the parameters of the display screen are adjusted in real time, so that the naked-eye 3D content can present the optimal visual effect under different illumination conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of naked-eye 3D display, and in particular to a naked-eye 3D interactive display processing method applied in a giant spherical building. Background Art

[0002] With the continuous advancement of science and technology, display technology is changing with each passing day. With its unique visual effects, naked-eye 3D display technology has shown great application potential in many fields, such as advertising media, film and television entertainment, education and training, medical diagnosis, virtual reality, etc. It allows users to directly watch images with a sense of three-dimensionality and depth without the help of any auxiliary equipment, bringing people an unprecedented visual experience.

[0003] However, in a special space like a giant ball building, how to achieve high-quality naked-eye 3D interactive display and provide users with a rich and natural interactive experience is still a technical problem that needs to be solved. Therefore, it is necessary to provide a naked-eye 3D interactive display processing method applied to a giant ball building to solve the above technical problems. Summary of the invention

[0004] The present invention provides a naked-eye 3D interactive display processing method applied in a giant spherical building, which solves the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides a naked-eye 3D interactive display processing method for a giant spherical building, comprising the following steps: S1: Identify the display screen group inside the giant ball building and collect spatial information to generate a spatial three-dimensional model; generate naked-eye 3D content based on the spatial three-dimensional model, and map the naked-eye 3D content on the display screen group; S2: Identify selected interactive users; S3: Establishing an interactive mechanism, including a hand recognition mechanism and a voice recognition mechanism; setting an operation identification library for any mechanism in the interactive mechanism, including predefined gesture execution instructions and response voice commands; S4: Select at least one interactive mechanism, identify and analyze the interactive information corresponding to the interactive mechanism, specifically: S41: For the gesture recognition mechanism, analyzing the gesture information of the selected interactive user to obtain a predefined gesture execution instruction; S42: For the voice recognition mechanism, natural language processing technology is used to analyze the voice information of the selected interactive user to obtain a response voice instruction; S5: Execute instructions according to gestures or respond to voice instructions to control corresponding contents in the naked-eye 3D contents to perform corresponding operations.

[0006] Preferably, the gesture information of the selected interactive user is analyzed, and the specific steps are as follows: Obtaining a hand video of a selected interactive user, and dividing the hand video into a number of hand time periods in chronological order; An infrared thermal image of the hand at any moment in the video is obtained through an infrared sensor, and the hand contour is obtained by processing the infrared thermal image through edge recognition technology; Obtain the hand contour at any moment within the hand time period, calculate the similarity between the hand contour at any moment and the hand contour at the previous moment, and obtain the hand change value; calculate the standard deviation of all hand change values ​​within the hand time period to obtain the hand fluctuation value; set the hand fluctuation threshold, if the hand fluctuation threshold is greater than the hand fluctuation threshold, mark the hand time period as a hand action time zone; connect adjacent hand action time zones to obtain a hand dynamic time zone; identify the time length of the hand dynamic time zone, set an extremely short time length threshold, compare the time length of the hand dynamic time zone with the extremely short time length threshold, if the time length of the hand dynamic time zone is less than the extremely short time length threshold, unmark the hand dynamic time zone; Obtain the gesture action corresponding to the gesture execution instruction predefined in the operation identification library of the gesture mechanism, and identify the preset gesture contour contained in the gesture action; perform similarity calculations on the preset gesture contours and the hand contours in the hand dynamic time zone in time sequence to obtain a gesture similarity value; if the gesture similarity value is greater than the set similarity threshold, it indicates that the gesture is standard, and perform similarity calculations on the next preset gesture contour and the hand contour at the next moment in the hand dynamic time zone; if the gesture similarity value is less than the set similarity threshold, it indicates that the gesture is not standard, and perform similarity calculations on the preset gesture contour and the hand contour at the next moment in the hand dynamic time zone until the gesture similarity value is greater than the set similarity threshold; The gesture similarity values ​​corresponding to all preset gesture contours are weightedly calculated to obtain a comprehensive gesture similarity value; the gesture comprehensive similarity values ​​corresponding to all predefined gestures are arranged to obtain a gesture similarity list; the predefined gesture corresponding to the largest gesture comprehensive similarity value in the gesture similarity list is selected as a pre-similar gesture; a gesture threshold is set, and the gesture comprehensive similarity value of the pre-similar gesture is compared with its gesture threshold. If the gesture comprehensive similarity value is greater than its gesture threshold, a predefined gesture execution instruction is generated.

[0007] Preferably, natural language processing technology is used to analyze the voice information of the selected interactive user, specifically: Collect the voice information of the selected interactive user, identify the voice signal in the voice information and perform pre-emphasis processing on the voice signal; divide the voice signal into several frames of short time periods, and apply a window function to each frame of the voice signal; Extract acoustic parameters reflecting speech characteristics from each frame of speech signal, the acoustic parameters including Mel-frequency cepstrum coefficients, linear prediction coefficients, fundamental frequency, and short-time energy; set a time window, calculate the mean and standard deviation of any parameter value in the acoustic parameters within the time window, and obtain the acoustic parameter mean and acoustic parameter fluctuation value; perform weighted calculation on the parameter values ​​in the acoustic parameters and the corresponding acoustic parameter mean and acoustic parameter fluctuation value to obtain the acoustic parameter influence value; perform weighted calculation on the acoustic parameter influence values ​​of all parameters in the acoustic parameters to obtain the acoustic feature vector; A hidden Markov model is set, which includes multiple states, each state represents a specific speech feature pattern, and an initial probability distribution is defined to represent the probability of the model being in each state at the initial moment, and a transition probability matrix is ​​used to describe the transition probability between states; The acoustic feature vector is input into the trained hidden Markov model, and the state sequence that is most likely to produce the observation sequence is searched in the hidden Markov model using the Viterbi algorithm decoding method. The state sequence is the recognition result of the speech signal. According to the speech units corresponding to the obtained state sequence, combined with the operation identification library and grammar rules corresponding to the pre-built language mechanism, the response voice command is parsed.

[0008] Preferably, based on step S1, the step of identifying the display screen group inside the giant spherical building and collecting spatial information to generate a spatial three-dimensional model further includes using laser scanning technology to accurately measure the spatial dimensions inside the building.

[0009] Preferably, it also includes a user interface for displaying in real time the changes in naked-eye 3D content corresponding to the user's gesture execution instructions or the response to voice instructions; and for providing operation guidance for the voice recognition mechanism or gesture recognition mechanism, the operation guidance includes graphics, text or animation form.

[0010] Preferably, based on step S3, the interaction mechanism further includes an ambient light adaptation mechanism, specifically: obtaining the light intensity in the giant ball building; The giant spherical building is divided into several spherical areas, and the mean and standard deviation of the light intensity at any position in the spherical area are calculated to obtain the light intensity mean and light intensity zone fluctuation value; Calculate the distance between the position of the area inside the ball and the position of any display screen in the display screen group inside the giant ball building to obtain the ball shadow distance value; The regional impact value is obtained by weighted calculation of the light intensity mean value, light intensity zone fluctuation value, and spherical shadow distance value; The regional influence values ​​of all the areas within the sphere are weighted and calculated to obtain the adaptation adjustment value corresponding to the display screen; the adaptation adjustment value is used to fine-tune and compensate the brightness and contrast of the display screen.

[0011] Preferably, it also includes a dynamically updated content library, including various 3D models, animations, videos and interactive applications for naked-eye 3D display.

[0012] According to the interactive display processing method for naked-eye 3D in a giant spherical building as described in claim 1, based on step S2, when identifying the selected interactive user, if no valid selected interactive user is detected within a preset time, it automatically switches to the demonstration mode and plays the preset naked-eye 3D content and the corresponding interactive effect examples in a loop.

[0013] Compared with the related art, the naked eye 3D interactive display processing method applied to a giant spherical building provided by the present invention has the following beneficial effects: 1. The present invention establishes an interactive mechanism, through hand recognition and voice recognition mechanisms, so that users can naturally interact with naked-eye 3D content, and combines the ambient light adaptation mechanism to perceive changes in ambient light and adjust the parameters of the display screen in real time to ensure that naked-eye 3D content can present the best visual effect under different lighting conditions.

[0014] 2. The present invention analyzes the light intensity at different positions in the giant spherical building through the ambient light adaptation mechanism, obtains the light intensity mean, light intensity fluctuation value and spherical shadow distance value and performs comprehensive calculations to obtain the adaptation adjustment value to fine-tune the brightness and contrast of the display screen, so that the naked-eye 3D content can maintain clear and comfortable visual effects under different lighting conditions, avoiding the problem of visual discomfort or poor display effect caused by lighting changes, and improving the user's viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flowchart of the interactive display processing method of naked-eye 3D applied in a giant spherical building provided by the present invention. DETAILED DESCRIPTION

[0016] 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.

[0017] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. The singular forms of "group", "class" and "the" used in this disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0018] It should be understood that although the terms first, second, third, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present disclosure, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0019] Please refer to Figure 1 The method for processing naked-eye 3D interactive display applied to a giant spherical building comprises the following steps: S1: Identify the display screen group inside the giant ball building and collect spatial information to generate a spatial three-dimensional model; generate naked-eye 3D content based on the spatial three-dimensional model, and map the naked-eye 3D content on the display screen group; S2: Identify selected interactive users; S3: Establishing an interactive mechanism, including a hand recognition mechanism and a voice recognition mechanism; setting an operation identification library for any mechanism in the interactive mechanism, including predefined gesture execution instructions and response voice commands; S4: Select at least one interactive mechanism, identify and analyze the interactive information corresponding to the interactive mechanism, specifically: S41: For the gesture recognition mechanism, analyzing the gesture information of the selected interactive user to obtain a predefined gesture execution instruction; S42: For the voice recognition mechanism, natural language processing technology is used to analyze the voice information of the selected interactive user to obtain a response voice instruction; S5: Execute instructions according to gestures or respond to voice instructions to control corresponding contents in the naked-eye 3D contents to perform corresponding operations.

[0020] Analyze the gesture information of the selected interactive user. The specific steps are as follows: Obtain hand videos of the selected interactive user from multiple angles, and divide the hand videos from any angle into several hand time periods in chronological order; An infrared thermal image of the hand at any moment in the video is obtained through an infrared sensor, and the hand contour is obtained by processing the infrared thermal image through edge recognition technology; Obtain the hand contour at any moment within the hand time period, calculate the similarity between the hand contour at any moment and the hand contour at the previous moment, and obtain the hand change value; calculate the standard deviation of all hand change values ​​within the hand time period to obtain the hand fluctuation value; set the hand fluctuation threshold, if the hand fluctuation threshold is greater than the hand fluctuation threshold, mark the hand time period as a hand action time zone; connect adjacent hand action time zones to obtain a hand dynamic time zone; identify the time length of the hand dynamic time zone, set an extremely short time length threshold, compare the time length of the hand dynamic time zone with the extremely short time length threshold, if the time length of the hand dynamic time zone is less than the extremely short time length threshold, unmark the hand dynamic time zone; Obtain the gesture action corresponding to the gesture execution instruction predefined in the operation identification library of the gesture mechanism, and identify the preset gesture contour contained in the gesture action; perform similarity calculations on the preset gesture contours and the hand contours in the hand dynamic time zone in chronological order to obtain a gesture similarity value XS; if the gesture similarity value is greater than the set similarity threshold, it indicates that the gesture is standard, and the next preset gesture contour is calculated to have a similarity with the hand contour at the next moment in the hand dynamic time zone; if the gesture similarity value is less than the set similarity threshold, it indicates that the gesture is not standard, and the preset gesture contour is calculated to have a similarity with the hand contour at the next moment in the hand dynamic time zone until the gesture similarity value is greater than the set similarity threshold; The gesture similarity values ​​corresponding to all preset gesture contours are weighted and calculated to obtain the gesture comprehensive similarity value ZX, which is expressed by the formula: ; Wherein, z represents the number of the preset gesture profile contained in the gesture action corresponding to the predefined gesture execution instruction, Z represents the total number of preset gesture profiles contained in the gesture action, and zXS and zq represent the gesture similarity value and its corresponding weight of the preset gesture profile numbered z in the gesture action; Obtain the comprehensive similarity value of gestures corresponding to predefined gesture actions in gesture videos at any angle; The gesture comprehensive similarity values ​​corresponding to the same predefined gesture action in all angle gesture videos are weighted to obtain the multi-angle gesture comprehensive similarity value J corresponding to the gesture action, which is expressed by the formula: ; Wherein, i represents the number of the acquisition angle of the multi-angle hand video, W represents the total number of angles collected by the hand video, iXS, ib represent the gesture comprehensive similarity value of the gesture action corresponding to the acquisition angle of the hand video at the number i and its corresponding weight respectively; Arrange the comprehensive similarity values ​​of all predefined gesture actions in order of size to obtain a gesture similarity list; select the predefined gesture action corresponding to the largest comprehensive similarity value of the gesture in the gesture similarity list and mark it as a pre-similar gesture; set the gesture threshold, and compare the comprehensive similarity value of the multi-angle gesture of the pre-similar gesture with its gesture threshold. If the comprehensive similarity value of the multi-angle gesture is greater than its gesture threshold, generate a predefined gesture execution instruction; if the comprehensive similarity value of the multi-angle gesture is less than or equal to its gesture threshold.

[0021] Use natural language processing technology to analyze the voice information of selected interactive users, specifically: Collect the voice information of the selected interactive user, identify the voice signal in the voice information and perform pre-emphasis processing on the voice signal; divide the voice signal into several frames of short time periods, and apply a window function to each frame of the voice signal; Extract acoustic parameters reflecting speech characteristics from each frame of speech signal, including Mel frequency cepstrum coefficients, linear prediction coefficients, and short-time energy; set a time window, calculate the mean and standard deviation of any parameter value in the acoustic parameters within the time window, and obtain the acoustic parameter mean DF2 and acoustic parameter fluctuation value DF3; identify the parameter value in the acoustic parameter as DF1, and perform weighted calculation on the parameter value in the acoustic parameter and its corresponding acoustic parameter mean and acoustic parameter fluctuation value to obtain the acoustic parameter influence value DF, which is expressed by the formula: ; Among them, r1, r2, and r3 represent the weights corresponding to the parameter value, the mean value of the acoustic parameter, and the fluctuation value of the acoustic parameter in the acoustic parameter respectively; the acoustic parameter influence values ​​of all parameters in the acoustic parameter are weighted and calculated to obtain the acoustic characteristic vector D, which is expressed by the formula: ; Wherein, t represents the number of the parameter in the acoustic parameters, H represents the total number of parameters in the acoustic parameters, and tδ represents the weight corresponding to the parameter t in the acoustic parameters; A hidden Markov model is set, which includes multiple states, each state represents a specific speech feature pattern, and an initial probability distribution is defined to represent the probability of the model being in each state at the initial moment, and a transition probability matrix is ​​used to describe the transition probability between states; The acoustic feature vector is input into the trained hidden Markov model, and the Viterbi algorithm decoding method is used to search the state sequence that is most likely to produce the observation sequence in the hidden Markov model. This state sequence is the recognition result of the speech signal; according to the speech units corresponding to the obtained state sequence, combined with the operation identification library and grammar rules corresponding to the pre-built language mechanism, the response voice command is parsed.

[0022] The specific steps of extracting acoustic parameters reflecting speech characteristics are as follows: Among them, the specific steps of extracting Mel frequency cepstrum coefficients are: Perform fast Fourier transform on each frame of windowed speech signal to obtain the spectrum of the speech signal, and convert the time domain signal of the speech signal into a frequency domain signal; A set of Mel filter banks is set based on the nonlinear frequency perception characteristics of the human auditory system; The actual frequency of the identified speech signal is represented by f, and the Mel frequency is calculated based on the actual frequency. The formula is: ; Among them, 700 represents the reference frequency value of the nonlinear perception characteristics of the human auditory system for different frequencies, 2595 represents the proportional coefficient for converting Mel frequency, log 10 Indicates logarithmic operation with base 10; Passing the frequency domain signal through a set of Mel filter banks, the Mel filter bank includes a plurality of triangular filters; Compute the response of each triangular filter in the frequency domain signal , the formula is: ; Wherein, m represents the number of the triangular filter in the Mel filter bank, k represents the frequency index, f(m) represents the center frequency of the triangular filter, and f(m-1) and f(m+1) represent the center frequencies of the previous and next triangular filters of the triangular filter numbered m, respectively. The logarithmic energy is calculated using the response of each triangular filter in the frequency domain signal to obtain the logarithmic Mel spectrum. , the formula is: ; X(k) represents the frequency domain signal after fast Fourier transform, represents the response of the mth Mel filter, k=0, 1..., N-1, N represents the total number of sampling points contained in the frame of the speech signal; It should be noted that N-1 ensures that all frequency domain sampling points are multiplied and summed with the responses of the corresponding Mel filters, thereby obtaining the energy contribution of each triangular filter in the entire frequency domain signal range; Then perform discrete cosine transform on the logarithmic Mel spectrum to obtain the Mel frequency cepstrum coefficient , the formula is: ; Wherein, T represents the serial number of the cepstral coefficient, and M represents the total number of triangular filters in the Mel filter bank; Among them, the specific steps of extracting the linear prediction coefficient are: Get the frames and application window functions contained in each frame of speech signal, and calculate the autocorrelation function of each frame of speech signal , the formula is: ; Wherein, N represents the total number of sampling points contained in the frame of the speech signal, x[g] represents the value of the g-th sampling point in the frame of the speech signal, and x[g+e], e represents the set time delay; According to the autocorrelation function, the linear prediction coefficients are obtained by solving the linear equations, and a p-order linear prediction model is constructed, which is expressed as: ; Where av represents the vth initial linear prediction coefficient, and p represents the prediction order; The prediction error signal is calculated using the initial linear prediction coefficients. The formula is: ; Wherein, x(n) represents the value of the nth sampling point in the speech signal of the frame, v represents the serial number of the initial linear prediction coefficient, and av represents the vth initial linear prediction coefficient; a prediction error threshold is set. If the prediction error signal is greater than the prediction error threshold, it means that the current initial linear prediction model is not accurate enough, and an adjustment signal is generated; the adjustment signal is used to increase the set prediction order or re-estimate the linear prediction coefficient; if the prediction error signal is less than or equal to the prediction error threshold, it means that the error is within the allowable range; Then the initial linear prediction coefficient is normalized, and the linear prediction coefficient is obtained by normalization calculation using the minimum mean square error criterion; Among them, the specific steps of extracting short-time energy are:

[0023] Calculate the short-time energy E of each frame of speech signal, the formula is: ; Where N represents the total number of sampling points contained in the frame of the speech signal, and x(n) represents the value of the nth sampling point in the frame of the speech signal; Based on step S1, in the step of identifying the display screen group inside the giant ball building and collecting spatial information to generate a spatial three-dimensional model, it further includes using laser scanning technology to accurately measure the spatial dimensions inside the building; by using laser scanning technology with high-precision measurement capabilities, it is possible to quickly and accurately obtain spatial information of various positions inside the giant ball building, including length, width, height and the relative position relationship between different display screens. The laser scanning technology improves the accuracy of spatial dimension measurement and lays a solid foundation for generating an accurate spatial three-dimensional model. At the same time, the laser scanning technology can scan a large area of ​​space in a short time, thereby improving the efficiency of data collection and saving time and labor costs.

[0024] The present invention also includes a user interface for displaying in real time the changes in naked-eye 3D content corresponding to the user's gesture execution instructions or the response to voice instructions; and for providing operation guidance for the voice recognition mechanism or the gesture recognition mechanism, the operation guidance including graphics, text or animation forms; by providing the operation guidance of the voice recognition mechanism or the gesture recognition mechanism, it is presented in the form of graphics, text or animation, etc., providing users with clear and easy-to-understand operation instructions, which is especially important for first-time users, and can help them get started quickly, reduce learning costs and operation difficulties. At the same time, the diversified forms of operation guidance can also meet the learning habits and needs of different users, thereby improving ease of use and versatility.

[0025] Based on step S3, the interactive mechanism also includes an ambient light adaptation mechanism, specifically: obtaining the light intensity inside the giant ball building; The giant spherical building is divided into several spherical areas, and the mean and standard deviation of the light intensity at any position in the spherical area are calculated to obtain the light intensity mean GZ1 and the light intensity area fluctuation value GZ2;

[0026] Calculate the distance between the position of the area inside the ball and the position of any display screen in the display screen group inside the giant ball building to obtain the ball shadow distance value GZ3; The regional influence value GZ is obtained by weighted calculation of the light intensity mean value, light intensity zone fluctuation value, and spherical shadow distance value. The formula is: ; Among them, h1, h2, and h3 represent the weights corresponding to the light intensity mean, light intensity fluctuation value, and spherical shadow distance value respectively; The regional influence values ​​of all areas within the sphere are weighted and calculated, and the formula is expressed as , the adaptation adjustment value SY corresponding to the display screen is obtained; wherein, s represents the number of the intra-spherical area, B represents the total number of intra-spherical areas in the giant spherical building, sGZ and sβ represent the regional influence value of the display screen on the intra-spherical area numbered s in the giant spherical building and its corresponding weight respectively; the adaptation adjustment value is used to fine-tune and compensate the brightness and contrast of the display screen; it should be noted that, through the ambient light adaptation mechanism, the light intensity at different positions in the giant spherical building is analyzed, the light intensity mean, the light intensity zone fluctuation value and the spherical shadow distance value are obtained and comprehensively calculated, and the adaptation adjustment value is obtained to fine-tune the brightness and contrast of the display screen, so that the naked-eye 3D content can maintain a clear and comfortable visual effect under different lighting conditions, avoiding the problem of visual discomfort or poor display effect caused by lighting changes, and improving the user's viewing experience.

[0027] The present invention also includes a dynamically updated content library including various 3D models, animations, videos and interactive applications for naked eye 3D display.

[0028] Based on step S2, when identifying the selected interactive user, if no valid selected interactive user is detected within the preset time, it will automatically switch to the demonstration mode to loop the preset naked eye 3D content and the corresponding interactive effect examples.

[0029] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed in this disclosure. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0030] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A naked-eye 3D interactive display processing method applied to a giant spherical building, characterized in that: The following steps are involved: S1: Identify the display screen group inside the giant ball building and collect spatial information to generate a spatial three-dimensional model; generate naked-eye 3D content based on the spatial three-dimensional model, and map the naked-eye 3D content on the display screen group; S2: Identify selected interactive users; S3: Establishing an interactive mechanism, including a hand recognition mechanism and a voice recognition mechanism; setting an operation identification library for any mechanism in the interactive mechanism, including predefined gesture execution instructions and response voice commands; S4: Select at least one interaction mechanism, identify and analyze the interaction information corresponding to the interaction mechanism, specifically: S41: For the gesture recognition mechanism, analyzing the gesture information of the selected interactive user to obtain a predefined gesture execution instruction; S42: For the voice recognition mechanism, natural language processing technology is used to analyze the voice information of the selected interactive user to obtain a response voice instruction; S5: Execute instructions according to gestures or respond to voice instructions to control corresponding contents in the naked-eye 3D contents to perform corresponding operations.

2. The method for interactive display of naked eye 3D in a giant spherical building according to claim 1, characterized in that: Analyze the gesture information of the selected interactive user. The specific steps are as follows: Obtaining a hand video of a selected interactive user, and dividing the hand video into a number of hand time periods in chronological order; An infrared thermal image of the hand at any moment in the video is obtained through an infrared sensor, and the hand contour is obtained by processing the infrared thermal image through edge recognition technology; Obtain the hand contour at any moment within the hand time period, calculate the similarity between the hand contour at any moment and the hand contour at the previous moment, and obtain the hand change value; calculate the standard deviation of all hand change values ​​within the hand time period to obtain the hand fluctuation value; set the hand fluctuation threshold, if the hand fluctuation threshold is greater than the hand fluctuation threshold, mark the hand time period as a hand action time zone; connect adjacent hand action time zones to obtain a hand dynamic time zone; identify the time length of the hand dynamic time zone, set an extremely short time length threshold, compare the time length of the hand dynamic time zone with the extremely short time length threshold, if the time length of the hand dynamic time zone is less than the extremely short time length threshold, unmark the hand dynamic time zone; Obtaining the gesture action corresponding to the gesture execution instruction predefined in the operation identification library of the gesture mechanism, and identifying the preset gesture outline contained in the gesture action; According to the time sequence of the preset gesture contours, similarity calculations are performed with the hand contours in the hand dynamic time zone to obtain a gesture similarity value; if the gesture similarity value is greater than the set similarity threshold, it indicates that the gesture is standard, and the next preset gesture contour is calculated to have a similarity with the hand contour at the next moment in the hand dynamic time zone; if the gesture similarity value is less than the set similarity threshold, it indicates that the gesture is not standard, and the preset gesture contour is calculated to have a similarity with the hand contour at the next moment in the hand dynamic time zone until the gesture similarity value is greater than the set similarity threshold; Performing weighted calculation on the gesture similarity values ​​corresponding to all preset gesture contours to obtain a gesture comprehensive similarity value; arranging the gesture comprehensive similarity values ​​corresponding to all predefined gestures to obtain a gesture similarity list; Selecting a predefined gesture corresponding to the largest gesture comprehensive similarity value in the gesture similarity list as a pre-similar gesture; A gesture threshold is set, and the gesture comprehensive similarity value of the pre-similar gesture is compared with its gesture threshold. If the gesture comprehensive similarity value is greater than its gesture threshold, a predefined gesture execution instruction is generated.

3. The method for interactive display of naked eye 3D in a giant spherical building according to claim 1, characterized in that: Use natural language processing technology to analyze the voice information of selected interactive users, specifically: Collect the voice information of the selected interactive user, identify the voice signal in the voice information and perform pre-emphasis processing on the voice signal; divide the voice signal into several frames of short time periods, and apply a window function to each frame of the voice signal; Extract acoustic parameters reflecting speech characteristics from each frame of speech signal, the acoustic parameters including Mel-frequency cepstrum coefficients, linear prediction coefficients, fundamental frequency, and short-time energy; set a time window, calculate the mean and standard deviation of any parameter value in the acoustic parameters within the time window, and obtain the acoustic parameter mean and acoustic parameter fluctuation value; perform weighted calculation on the parameter values ​​in the acoustic parameters and the corresponding acoustic parameter mean and acoustic parameter fluctuation value to obtain the acoustic parameter influence value; perform weighted calculation on the acoustic parameter influence values ​​of all parameters in the acoustic parameters to obtain the acoustic feature vector; A hidden Markov model is set, which includes multiple states, each state represents a specific speech feature pattern, and an initial probability distribution is defined to represent the probability of the model being in each state at the initial moment, and a transition probability matrix is ​​used to describe the transition probability between states; The acoustic feature vector is input into the trained hidden Markov model, and the state sequence that is most likely to produce the observation sequence is searched in the hidden Markov model using the Viterbi algorithm decoding method. The state sequence is the recognition result of the speech signal. According to the speech units corresponding to the obtained state sequence, combined with the operation identification library and grammar rules corresponding to the pre-built language mechanism, the response voice command is parsed.

4. The method for interactive display of naked eye 3D in a giant spherical building according to claim 1, characterized in that: Based on step S1, the step of identifying the display screen group inside the giant spherical building and collecting spatial information to generate a spatial three-dimensional model further includes using laser scanning technology to accurately measure the spatial dimensions inside the building.

5. The method for interactive display of naked eye 3D in a giant spherical building according to claim 1, characterized in that: It also includes a user interface for displaying in real time the changes in naked-eye 3D content corresponding to the user's gesture execution instructions or response to voice instructions; it is also used to provide operation guidance for the voice recognition mechanism or gesture recognition mechanism, and the operation guidance includes graphics, text or animation form.

6. The method for interactive display of naked eye 3D in a giant spherical building according to claim 1, characterized in that: Based on step S3, the interaction mechanism also includes an ambient light adaptation mechanism, specifically: obtaining the light intensity inside the giant ball building; The giant spherical building is divided into several spherical areas, and the mean and standard deviation of the light intensity at any position in the spherical area are calculated to obtain the light intensity mean and light intensity zone fluctuation value; Calculate the distance between the position of the area inside the ball and the position of any display screen in the display screen group inside the giant ball building to obtain the ball shadow distance value; The regional impact value is obtained by weighted calculation of the light intensity mean value, light intensity zone fluctuation value, and spherical shadow distance value; Perform weighted calculation on the regional influence values ​​of all areas within the sphere to obtain the corresponding adaptation adjustment value of the display screen; Adaptive adjustment values ​​are used to fine-tune and compensate for the brightness and contrast of the display.

7. The method for interactive display of naked eye 3D in a giant spherical building according to claim 1, characterized in that: It also includes a dynamically updated content library of various 3D models, animations, videos and interactive applications for naked-eye 3D display.

8. According to claim 1, the method for interactive display processing of naked-eye 3D in a giant spherical building, based on step S2, when identifying the selected interactive user, if no valid selected interactive user is detected within a preset time, it automatically switches to the demonstration mode and plays the preset naked-eye 3D content and the corresponding interactive effect examples in a loop.

Images