A method and system for realizing interactive lighting interaction control

Abstract

The present invention provides a method and system for realizing interactive lighting interaction control, which relates to the technical field of lighting interaction design. The purpose is to achieve more targeted and richer lighting interaction displays by considering more comprehensive population characteristics, including: sampling the position coordinates of each interaction target at equal time intervals within the first time period; predicting the predicted position coordinates of each interaction target at the mid-time point of the second time period based on the first model; respectively sampling the facial orientation of each interaction target at each position coordinate and the distance factor relative to each lighting point; obtaining the lighting interaction type probability parameters based on the second model according to the facial orientation array and the distance factor; respectively obtaining all the adapted interaction targets of each lighting point at the mid-time point of the second time period; and obtaining the interaction lighting type of the lighting point in the second time period. The present invention has the advantages of more flexible and comprehensive lighting interaction control.

Description

Technical Field

[0001] The present invention relates to the technical field of lighting interaction design, and more particularly, to a lighting interaction control method and system for realizing interaction. Background Art

[0002] Lighting interaction displays have gradually become a common decorative setting in various exhibition halls and other display or entertainment venues.

[0003] In order to achieve better lighting interaction displays, it is necessary to perform more targeted control of lighting interaction based on the characteristics of the crowd. In the prior art, lighting interaction can be controlled according to the crowd gathering characteristics, or the lighting can be adjusted only according to the orientation position of each person. However, these forms of control are relatively single and are not conducive to comprehensively and richly performing more targeted lighting interaction displays.

[0004] Therefore, it is necessary to optimize the lighting interaction control method to achieve more targeted and richer lighting interaction displays by considering more comprehensive crowd characteristics. Summary of the Invention

[0005] The purpose of the present invention is to provide a lighting interaction control method and system for realizing interaction, which can achieve more targeted and richer lighting interaction displays by considering more comprehensive crowd characteristics.

[0006] The embodiments of the present invention are implemented through the following technical solutions:

[0007] The present invention first provides a lighting interaction control method for realizing interaction, including the following steps:

[0008] Sampling the position coordinates of each interaction target at equal time intervals within a first time period to obtain a position coordinate array;

[0009] Based on a first model, respectively predicting the predicted position coordinates of each interaction target at the mid-time point of a second time period, where the second time period is the next time period of the first time period;

[0010] Sampling the facial orientation of each interaction target at each position coordinate and the distance factor relative to each lighting point to obtain a facial orientation array and a distance factor array;

[0011] Based on a second model, respectively obtaining the lighting interaction type probability parameter clas ij clas ij represents the probability that the recommended lighting interaction type of the i-th interaction target in the second time period is the lighting interaction type of the j-th lighting point in the first time period;

[0012] Obtain all the adapted interaction targets at the mid-time point of each lighting point in the second time period respectively;

[0013] Respectively according to the lighting interaction type probability parameter clas of all the adapted interaction targets of the lighting point ij Obtain the interaction lighting type of the lighting point in the second time period.

[0014] Preferably, the first model adopts a recurrent neural network or a long short-term memory network.

[0015] Preferably, the second model includes an input layer, a first feature extraction layer, a second feature extraction layer, a splicing layer, and a classification output layer;

[0016] The input layer is used to receive the facial orientation array and the distance factor array;

[0017] The first feature extraction layer is used to extract a first feature matrix according to the facial orientation array;

[0018] The second feature extraction layer is used to extract a second feature matrix according to the first feature matrix and the distance factor array;

[0019] The splicing layer is used to splice the first feature matrix and the second feature matrix to obtain a fused feature matrix;

[0020] The classification output layer is used to output the lighting interaction type probability parameter clas according to the fused feature matrix ij 。

[0021] Preferably, the expression of the first feature extraction layer is:

[0022]

[0023] Where, SP i 1 is the first feature matrix of the i-th interaction target, N is the total number of lighting points, is the j-th element in SP i 1 m represents the m-th sampling of the data used in the first time period, M is the total number of sampled data in the first time period, e is the natural constant, ln(.) is the natural logarithm function, min j (.,.) is a function that takes the minimum value based on the change of j, δ is a preset angle difference threshold, ag ij,m represents the included angle between the vector pointing from the i-th interaction target to the j-th lighting point and the facial orientation of the i-th interaction target at the m-th sampling, round(.) represents the rounding function, char ij,mRepresents the intermediate parameter from the i-th interaction target to the j-th lighting point during the m-th sampling. If is a truth judgment function and returns 1 when the judgment object is true, otherwise returns 0.

[0024] Preferably, the expression of the second feature extraction layer is:

[0025]

[0026] Where, SP i 2 Is the second feature matrix of the i-th interaction target, N is the total number of lighting points, Is SP i 2 The j-th element in, d ij,m Is the distance factor from the i-th interaction target to the j-th lighting point during the m-th sampling, L iq,m And L ij,m Are respectively the distances from the i-th interaction target to the q-th and j-th lighting points during the m-th sampling, L th Is a preset distance threshold.

[0027] Preferably, the expression of the splicing layer is:

[0028]

[0029] Where, SP i mix Is the fusion feature matrix.

[0030] Preferably, the expression of the classification output layer is:

[0031]

[0032] α + β = 1;

[0033] Where, both α and β are weights to be trained.

[0034] Preferably, the method for respectively obtaining all the adapted interaction targets at the intermediate time points of each lighting point in the second time period is:

[0035] Respectively obtain the distances from the interaction target to each lighting point based on the predicted position coordinates of the interaction target at the intermediate time point in the second time period;

[0036] Obtain the lighting point closest to the interaction target, and this interaction target is the adapted interaction target corresponding to the lighting point.

[0037] Preferably, according to the lighting interaction type probability parameters clas of all the adapted interaction targets of the lighting point positions respectively ij The method for obtaining the interaction lighting type of the lighting point position in the second time period is as follows:

[0038] Obtain the lighting interaction type probability parameters clas of each of the adapted interaction targets respectively ij ;

[0039] Calculate the evaluation parameters of each of the adapted interaction targets respectively:

[0040]

[0041] where evp k is the evaluation parameter of the k-th adapted interaction target of the lighting point position, and i k represents the serial number of the k-th adapted interaction target among all the interaction targets, that is is the probability that the recommended lighting interaction type of the i k -th interaction target in the second time period is the lighting interaction type of the j-th lighting point position in the first time period, e is the natural constant, N is the total number of the lighting point positions, and am k is the number of values in the lighting interaction type probability parameter of the k-th adapted interaction target that are greater than the probability parameter threshold SUM max3 is the sum of the three largest values in the lighting interaction type probability parameter of the k-th adapted interaction target;

[0042] Obtain the correction weight based on the evaluation parameter:

[0043]

[0044] where corre k is the correction coefficient of the k-th adapted interaction target of the lighting point position, V is the total number of the adapted interaction targets of the lighting point position, and evp v is the evaluation parameter of the v-th adapted interaction target of the lighting point position;

[0045] Obtain the probability comprehensive parameters of each of the interaction lighting types respectively:

[0046]

[0047] where J j represents the probability correction parameter that the recommended lighting interaction type in the second time period is the lighting interaction type of the j-th lighting point position in the first time period, and k is the total number of the adapted interaction targets;

[0048] Obtain the value of J with the maximum probability correction parameter jmax , where jmax ∈ [1, N], and the interactive lighting type of the lighting point in the second time period is the lighting interaction type of the jmax-th lighting point in the first time period

[0049] To solve the above problems, the present invention also provides an interactive lighting interaction control system, which is applied to the interactive lighting interaction control method described in any one of the above, and is characterized by including:

[0050] A first sampling module, configured to sample the position coordinates of each interaction target at equal time intervals in the first time period to obtain a position coordinate array

[0051] A prediction module, configured to predict the predicted position coordinates of each interaction target at the middle time point of the second time period based on the first model according to the position coordinate array, where the second time period is the next time period of the first time period

[0052] A second sampling module, configured to sample the facial orientation and distance factor relative to each lighting point of each interaction target at each position coordinate to obtain a facial orientation array and a distance factor array

[0053] A type recognition module, configured to obtain the lighting interaction type probability parameter clas based on the second model according to the facial orientation array and the distance factor ij , clas ij represents the probability that the recommended lighting interaction type of the i-th interaction target in the second time period is the lighting interaction type of the j-th lighting point in the first time period

[0054] An adaptive interaction target acquisition module, configured to respectively acquire all adaptive interaction targets of each lighting point at the middle time point of the second time period

[0055] An interactive lighting type judgment module, configured to respectively obtain the interactive lighting type of the lighting point in the second time period according to the lighting interaction type probability parameter clas of all adaptive interaction targets of the lighting point ij

[0056] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0057] The present invention can adjust the lighting periodically, with flexible adjustment, which is convenient for dealing with real-time changes. And each adjustment determines the lighting adjustment plan based on the crowd position and face orientation data of the next period deduced from the data of the previous period, making the lighting adjustment more targeted to the crowd characteristics

[0058] ​When the present invention extracts features through face orientation data, it simultaneously considers the observation tendency of people and the influence of distance factors on the observation tendency, thereby achieving more comprehensive feature consideration. When adjusting the light type, the pertinence and richness are simultaneously improved;

[0059] After the present invention obtains the comprehensive features of each person, it classifies them according to the distance from the light point, and assigns different weights to different people based on the characteristics of the observed population at the light point to obtain the final light adjustment type, further improving the comprehensiveness of the control considerations;

[0060] The present invention is reasonably designed and is convenient to be applied to various light display interaction scenarios, facilitating promotion and implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic structural diagram of a light interaction control method for realizing interaction provided in Embodiment 1 of the present invention;

[0062] Figure 2 It is a schematic structural diagram of a second model provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations.

[0064] Embodiment 1

[0065] This embodiment provides a light interaction control method for realizing interaction. Refer to Figure 1 , including the following steps:

[0066] Step S1: Sample the position coordinates of each interaction target at equal time intervals within the first time period to obtain a position coordinate array;

[0067] Step S2: Based on the first model, respectively predict the predicted position coordinates of each interaction target at the mid-time point of the second time period according to the position coordinate array, where the second time period is the next time period of the first time period;

[0068] Step S3: Sample the face orientation of each interaction target at each position coordinate and the distance factor relative to each light point to obtain a face orientation array and a distance factor array;

[0069] Step S4: Based on the second model, obtain the lighting interaction type probability parameters clas respectively according to the facial orientation array and the distance factor ij , clas ij represents the probability that the recommended lighting interaction type of the i-th interaction target in the second time period is the lighting interaction type of the j-th lighting point in the first time period;

[0070] Step S5: Obtain all the adapted interaction targets at the mid-time point of each lighting point in the second time period respectively;

[0071] Step S6: Obtain the interactive lighting type of the lighting point in the second time period respectively according to the lighting interaction type probability parameter clas ij of all the adapted interaction targets of the lighting point.

[0072] As a preferred solution of this embodiment, the first model in step S2 adopts a recurrent neural network or a long short-term memory network. Both of these networks are relatively mature state prediction networks and can be applied to predict the position state of the interaction target in this embodiment. It should be particularly noted that the facial orientation in step S3 can be estimated by existing deep learning models, such as OpenPose and MediaPipe. The sampling of the facial orientation and the position information can be realized by arranging camera devices in the scene.

[0073] This embodiment makes position predictions based on the sampling data of one time period, aiming to predict suitable audience interaction targets for each lighting point in the next time period, and at the same time extract the facial orientation array and the distance factor array to consider the most suitable lighting interaction type for each audience interaction target in the next time period from multiple aspects. Based on the above data acquisition, the lighting interaction type arrangement of each lighting point in the next time period is formulated, so as to achieve comprehensive and flexible lighting control.

[0074] Embodiment 2

[0075] This embodiment is based on the technical solution of Embodiment 1, and further illustrates the specific implementation methods of each step.

[0076] In this embodiment, referring to Figure 2 , the second model in step S4 includes an input layer, a first feature extraction layer, a second feature extraction layer, a splicing layer and a classification output layer;

[0077] The input layer is used to receive the facial orientation array and the distance factor array;

[0078] The first feature extraction layer is used to extract the first feature matrix according to the facial orientation array;

[0079] The second feature extraction layer is used to extract a second feature matrix according to the first feature matrix and the distance factor array;

[0080] The splicing layer is used to splice the first feature matrix and the second feature matrix to obtain a fused feature matrix;

[0081] The classification output layer is used to output the probability parameter clas of the lighting interaction type according to the fused feature matrix ij 。

[0082] First, the expression of the first feature extraction layer is preferably:

[0083]

[0084] where, SP i 1 is the first feature matrix of the i-th interaction target, N is the total number of the lighting points, is the j-th element in SP i 1 m represents the m-th sampling of the data adopted in the first time period, M is the total number of sampling data in the first time period, e is the natural constant, ln(.) is the natural logarithm function, min j (.,.) is a function that takes the minimum value based on the change of j, δ is a preset angle difference threshold, for example, it can be taken as 30 degrees, ag ij,m represents the included angle between the vector pointing from the i-th interaction target to the j-th lighting point and the facial orientation of the i-th interaction target at the m-th sampling, round(.) represents the rounding function, char ij,m represents the intermediate parameter of the i-th interaction target relative to the j-th lighting point at the m-th sampling, if is a truth judgment function and returns 1 when the judgment object is true and returns 0 otherwise.

[0085] The first feature extraction layer here is used to extract the facial orientation information, comprehensively considering the situation of the lighting points falling within a certain angle range of the facial orientation.

[0086] Further, the expression of the second feature extraction layer is preferably:

[0087]

[0088] where,, SP i 2 is the second feature matrix of the i-th interaction target, N is the total number of the lighting points, is the j-th element in SP i 2 d ij,mis the distance factor from the i-th interaction target to the j-th lighting point during the m-th sampling, L iq,m and L ij,m are the distances from the i-th interaction target to the q-th and j-th lighting points respectively during the m-th sampling, L th is a preset distance threshold. L iq,m and L ij,m are obtained by extracting from the distance factor array.

[0089] Preferably, the expression of the splicing layer is:

[0090]

[0091] where SP i mix is the fusion feature matrix.

[0092] As a further preferred solution, the expression of the classification output layer is:

[0093]

[0094] α + β = 1;

[0095] where α and β are both weights to be trained.

[0096] The purpose of adding the second feature extraction layer in this embodiment is to extract the distance information from the interaction target to each lighting point. Because the viewing tendency extracted from the facial orientation of the interaction panel will be affected by the distance and not simply by preferences, and in the case of a low preference at a close distance, the viewing tendency extracted from the facial orientation will also decrease, possibly because of the high viewing frequency due to the close distance. Therefore, this embodiment fuses two types of feature data, comprehensively considers the viewing tendency and distance factors, and tries to match as much as possible the types that cannot be seen due to too far a distance or have a particularly strong viewing intention. Therefore, this embodiment can achieve a more comprehensive feature consideration, and thus obtain a more detailed and highly matched lighting control scheme.

[0097] In this embodiment, the method for respectively obtaining all the adapted interaction targets of each lighting point at the middle time point of the second time period in step S5 is preferably:

[0098] Respectively obtain the distance from the interaction target to each lighting point based on the predicted position coordinates of the interaction target at the middle time point of the second time period;

[0099] Obtain the lighting point closest to the interaction target, and this interaction target is the adapted interaction target corresponding to the lighting point.

[0100] Finally, in step S6, according to the lighting interaction type probability parameters clas of all the adapted interaction targets at the lighting point positions ij The method for obtaining the interaction lighting type of the lighting point position in the second time period is as follows:

[0101] Obtain the lighting interaction type probability parameters clas of each of the adapted interaction targets respectively ij ;

[0102] Calculate the evaluation parameters of each of the adapted interaction targets respectively:

[0103]

[0104] where evp k is the evaluation parameter of the k-th adapted interaction target at the lighting point position, and i k represents the number of the k-th adapted interaction target among all the interaction targets, that is is the probability that the recommended lighting interaction type of the i k -th interaction target in the second time period is the lighting interaction type of the j-th lighting point position in the first time period, e is the natural constant, N is the total number of the lighting point positions, and am k is the number of values in the lighting interaction type probability parameter of the k-th adapted interaction target that are greater than the probability parameter threshold , and SUM max3 is the sum of the three largest values in the lighting interaction type probability parameter of the k-th adapted interaction target;

[0105] Obtain the correction weight based on the evaluation parameter:

[0106]

[0107] where corre k is the correction coefficient of the k-th adapted interaction target at the lighting point position, V is the total number of the adapted interaction targets at the lighting point position, and evp v is the evaluation parameter of the v-th adapted interaction target at the lighting point position;

[0108] Obtain the probability comprehensive parameters of each of the interaction lighting types respectively:

[0109]

[0110] where J j represents the probability correction parameter that the recommended lighting interaction type in the second time period is the lighting interaction type of the j-th lighting point position in the first time period, and k is the total number of the adapted interaction targets;

[0111] Obtain the value of J with the maximum probability correction parameter jmax , where jmax ∈ [1, N], and the interactive light type of the light point in the second time period is the light interaction type of the jmax-th light point in the first time period.

[0112] Finally, in this embodiment, the probability parameters of each light interaction type of each adapted interaction target at each light point are corrected, that is, different calculation weights are assigned to each adapted interaction target to compare which probability parameter of the light interaction type of this light point has the highest comprehensive evaluation. Assigning different calculation weights to each adapted interaction target mainly depends on the distribution of the predicted probabilities obtained for each adapted interaction target, that is, a higher weight is assigned to the adapted interaction target with a greater tendency. As a further optimization, when allocating the light interaction types of each light point in the second time period in this embodiment, the comprehensive coverage of types can also be considered, that is, the number of light points is set to be greater than or equal to the number of light interaction types, and when allocating, it is ensured as much as possible that each light interaction type appears. For example, if there is an unallocated light interaction type after the allocation, then find the light interaction type with the most allocations and replace the one with the lowest probability comprehensive parameter among them with the unallocated light interaction type.

[0113] Embodiment 3

[0114] This embodiment provides an interactive light interaction control system, which is applied to an interactive light interaction control method described in any one of the above, and is characterized by including:

[0115] A first sampling module, configured to sample the position coordinates of each interaction target at equal time intervals in the first time period to obtain a position coordinate array;

[0116] A prediction module, configured to respectively predict the predicted position coordinates of each interaction target at the middle time point of the second time period based on the first model according to the position coordinate array, where the second time period is the next time period of the first time period;

[0117] A second sampling module, configured to respectively sample the facial orientation of each interaction target at each position coordinate and the distance factor relative to each light point to obtain a facial orientation array and a distance factor array;

[0118] A type recognition module, configured to respectively obtain the light interaction type probability parameter clas based on the second model according to the facial orientation array and the distance factor ij , clas ijRepresents the probability that the recommended lighting interaction type of the \(i\)-th interaction target in the second time period is the lighting interaction type of the \(j\)-th lighting point in the first time period;

[0119] An adaptation interaction target acquisition module, configured to respectively acquire all adaptation interaction targets of each lighting point at the mid-time point of the second time period;

[0120] An interactive lighting type determination module, configured to respectively obtain the interactive lighting type of the lighting point in the second time period according to the lighting interaction type probability parameter clas of all adaptation interaction targets of the lighting point ij

[0121] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.​

Claims

1. An interactive lighting interaction control method, characterized in that, It includes the following steps: Sample the position coordinates of each interaction target at equal time intervals within the first time period to obtain a position coordinate array; Based on the first model, respectively predict the predicted position coordinates of each interaction target at the mid-time point of the second time period, where the second time period is the next time period after the first time period; Sample the facial orientation of each interaction target at each position coordinate and the distance relative to each light point position respectively to obtain a facial orientation array and a distance factor array; Based on the second model, obtain the lighting interaction type probability parameter clas according to the facial orientation array and the distance factor respectively ij , clas ij represents the probability that the recommended lighting interaction type of the i-th interaction target in the second time period is the lighting interaction type of the j-th lighting point position in the first time period; Respectively obtain all the adapted interaction targets of each light point position at the mid-time point of the second time period; According to the lighting interaction type probability parameters clas of all the adapted interaction targets of the lighting point positions respectively ij Obtain the interaction lighting type of the lighting point position in the second time period; The method for respectively obtaining all the adapted interaction targets of each light point position at the mid-time point of the second time period is: Respectively obtain the distance from the interaction target to each light point position based on the predicted position coordinates of the interaction target at the mid-time point of the second time period; Obtain the light point position closest to the interaction target, and this interaction target is the adapted interaction target corresponding to the light point position.

2. The method for realizing interactive lighting interaction control according to claim 1, wherein, The first model uses a recurrent neural network or a long short-term memory network.

3. The method for realizing interactive lighting interaction control according to claim 1, wherein The second model includes an input layer, a first feature extraction layer, a second feature extraction layer, a splicing layer, and a classification output layer; The input layer is used to receive the facial orientation array and the distance factor array; The first feature extraction layer is used to extract a first feature matrix according to the facial orientation array; The second feature extraction layer is used to extract a second feature matrix according to the first feature matrix and the distance factor array; The splicing layer is used to splice the first feature matrix and the second feature matrix to obtain a fused feature matrix; The classification output layer is used to output the probability parameter clas of the lighting interaction type according to the fusion feature matrix ij .

4. The method for realizing interactive lighting interaction control according to claim 3, wherein The expression of the first feature extraction layer is: Among them, SP i 1 is the first feature matrix of the i-th interaction target, and N is the total number of the lighting points. is SP i 1 is the j-th element in SP, m represents the m-th sampling of the data used within the first time period, M is the total number of sampling data within the first time period, e is the natural constant, ln(.) is the natural logarithm function, min j (.,.) is a function that takes the minimum value based on the change of j, and δ is a preset angle difference threshold, ag ij,m represents the included angle between the vector pointing from the i-th interaction target to the j-th lighting point and the face orientation of the i-th interaction target during the m-th sampling, round(.) represents the rounding function, char ij,m represents the intermediate parameter of the i-th interaction target relative to the j-th lighting point during the m-th sampling, if is a truth judgment function and returns 1 when the judgment object is true and otherwise returns 0.

5. The method for realizing interactive lighting interaction control according to claim 4, wherein The expression of the second feature extraction layer is: Among them, SP i 2 is the second feature matrix of the i-th interaction target, and N is the total number of the lighting points, is SP i 2 is the j-th element in, d ij,m is the distance factor from the i-th interaction target to the j-th lighting point at the m-th sampling, L iq,m and L ij,m are respectively the distances from the i-th interaction target to the q-th and j-th lighting points at the m-th sampling, L th is a preset distance threshold.

6. The method for realizing interactive lighting interaction control according to claim 5, wherein The expression of the splicing layer is: Among them, SP i mix is the fusion feature matrix.

7. The method for realizing interactive lighting interaction control according to claim 6, characterized in that, The expression of the classification output layer is: α+β＝1； Where α and β are both weights to be trained.

8. The method for realizing interactive lighting interaction control according to claim 1, wherein According to the lighting interaction type probability parameters clas of all the adapted interaction targets of the lighting point positions respectively ij The method for obtaining the interaction lighting type of the lighting point position in the second time period is as follows: Obtain the probability parameter clas of the lighting interaction type for each of the adaptation interaction targets respectively ij ; Calculate the evaluation parameters of each adapted interaction target respectively; Among them, evp k is the evaluation parameter of the k-th adaptation interaction target of the lighting point, i k represents the number of the k-th adaptation interaction target among all the interaction targets, that is is the probability that the recommended lighting interaction type of the i k -th interaction target in the second time period is the lighting interaction type of the j-th lighting point in the first time period. e is the natural constant, N is the total number of the lighting points, am k is the number of the lighting interaction type probability parameters of the k-th adaptation interaction target whose values are greater than the probability parameter threshold , SUM max3 is the sum of the three largest values among the lighting interaction type probability parameters of the k-th adaptation interaction target; Obtain the correction weight based on the evaluation parameters; Among them, corre k is the correction coefficient of the k-th adaptation interaction target of the lighting point, V is the total number of the adaptation interaction targets of the lighting point, evp v is the evaluation parameter of the v-th adaptation interaction target of the lighting point; Respectively obtain the probability comprehensive parameters of each interaction light type; Among them, J j represents the probability correction parameter that the recommended light interaction type in the second time period is the light interaction type of the j-th light point position in the first time period, and k is the total number of the adapted interaction targets; Obtain the value of J with the maximum probability correction parameter jmax , jmax ∈ [1, N], and the interactive lighting type of the lighting point in the second time period is the lighting interaction type of the jmax-th lighting point in the first time period.

9. An interactive lighting interaction control system, which is applied to an interactive lighting interaction control method described in any one of claims 1-8, and is characterized in that, It includes: A first sampling module for sampling the position coordinates of each interaction target at equal time intervals within the first time period to obtain a position coordinate array; A prediction module for predicting the predicted position coordinates of each interaction target at the mid-time point of the second time period based on the first model according to the position coordinate array, where the second time period is the next time period after the first time period; A second sampling module for sampling the facial orientation of each interaction target at each position coordinate and the distance factor relative to each light point position respectively to obtain a facial orientation array and a distance factor array; A type recognition module, configured to respectively obtain light interaction type probability parameters clas based on a second model according to a face orientation array and a distance factor ij , clas ij represents the probability that the recommended light interaction type of the i-th interaction target in the second time period is the light interaction type of the j-th light point position in the first time period; An adapted interaction target acquisition module for respectively obtaining all the adapted interaction targets of each light point position at the mid-time point of the second time period; The method for respectively obtaining all the adapted interaction targets of each light point position at the mid-time point of the second time period is: Respectively obtain the distance from the interaction target to each light point position based on the predicted position coordinates of the interaction target at the mid-time point of the second time period; Obtain the lighting point that is closest to the interaction target, where the interaction target is the adapted interaction target corresponding to the lighting point; The interactive lighting type judgment module is used to respectively obtain the interactive lighting type of the lighting point in the second time period according to the lighting interaction type probability parameter clas of all the adapted interaction targets of the lighting point. ij Obtain the interactive lighting type of the lighting point in the second time period.

Images