Virtual space generating apparatus, method, and program

Abstract

In one embodiment of the present invention, first, information representing a personality characteristic of a user and a target state of the user himself / herself required for a virtual space is acquired. On the basis of each set of acquired information, an initial value of a space parameter giving an impression matching the target state is determined, and a virtual space reflecting the set initial value of the space parameter is presented to the user US. Next, information representing the current state of the user US after the presentation of the virtual space is acquired, the space parameter is optimized so as to bring the acquired current state close to the target state, and a virtual space reflecting the optimized space parameter is generated and presented to the user US.

Description

Virtual space generation apparatus, method, and program

[0001] One aspect of this invention relates to a virtual space generation apparatus, method, and program used, for example, to generate a virtual space corresponding to the state of a user.

[0002] In recent years, technologies that use virtual reality to present various information to users have been put into practical use. By projecting 360-degree virtual space images into a person's field of view, it becomes possible to give users the feeling of actually being in a virtual space. However, designing a virtual space that suits the user's purpose requires a great deal of effort and time.

[0003] On the other hand, research investigating the relationship between user personality traits and their impressions of virtual space interiors has reported that user personality traits influence their perception of virtual space interiors (see, for example, Non-Patent Document 1). Applying this report, it becomes possible to generate virtual spaces that correspond to user personality traits by determining the desired impression based on the user's personality traits.

[0004] Banaei, Maryam, et al. “Emotional evaluation of architectural interior forms based on personality differences using virtual reality.” Frontiers of Architectural Research 9.1 (2020): 138-147.

[0005] However, in the technology that applies Non-Patent Document 1, while it is possible to determine the impression to be given to the user, there is often a discrepancy between the determined impression and the user's actual state, making it difficult to generate a virtual space that reflects the user's actual state.

[0006] This invention was made in view of the above circumstances and aims to provide a technology that makes it possible to generate a virtual space that reflects the user's actual state.

[0007] To solve the above problems, one embodiment of the virtual space generation device or generation method according to the present invention acquires information representing the user's personality traits and information representing the user's target state in the virtual space, and determines spatial parameters corresponding to an impression that matches the target state based on the acquired personality traits and target state. Then, with the virtual space information reflecting the spatial parameters presented to the user, information representing the user's current state is acquired, and the spatial parameters are optimized to bring the acquired current state closer to the target state.

[0008] According to one aspect of this invention, for example, even if the impression received from the virtual space reflecting the initial values ​​of the spatial parameters deviates from the target state desired by the user, the user's current state is acquired, and the spatial parameters are optimized so that this current state approaches the user's own target state sought in the virtual space. A virtual space reflecting the optimized spatial parameters is then generated. As a result, a virtual space reflecting the user's actual state can be generated and presented to the user.

[0009] In other words, according to one aspect of this invention, it is possible to provide a technology that enables the generation of a virtual space that reflects the actual state of the user.

[0010] Figure 1 is a diagram showing an example of the configuration of a virtual space presentation system equipped with a virtual space generation device according to the first embodiment of the present invention. Figure 2 is a block diagram showing an example of the hardware configuration of a virtual space generation device according to the first embodiment of the present invention. Figure 3 is a block diagram showing an example of the software configuration of a virtual space generation device according to the first embodiment of the present invention. Figure 4 is a flowchart showing an example of the processing procedure and processing content of a virtual space generation process executed by the control unit of the virtual space generation device shown in Figure 3. Figure 5 is a flowchart showing an example of the processing procedure and processing content of the spatial parameter determination process among the processing procedures shown in Figure 4. Figure 6 is a flowchart showing an example of the processing procedure and processing content of the spatial parameter optimization process among the processing procedures shown in Figure 4. Figure 7 is a block diagram showing an example of the software configuration of a virtual space generation device according to the second embodiment of the present invention. Figure 8 is a flowchart showing an example of the processing procedure and processing content of the spatial parameter determination process among the processing procedures of a virtual space generation process executed by the control unit of the virtual space generation device shown in Figure 7. Figure 9 is a flowchart showing an example of the processing procedure and processing content of the spatial parameter optimization process among the processing procedures of a virtual space generation process executed by the control unit of the virtual space generation device shown in Figure 7. Figure 10 is a diagram showing an example of a virtual space. Figure 11 is a table showing the results of the evaluation function obtained in the spatial parameter optimization process shown in Figure 9. Figure 12 is a diagram showing an example of the optimized spatial parameters obtained in the spatial parameter optimization process shown in Figure 9.

[0011] Embodiments of this invention will be described below with reference to the drawings.

[0012] [First Embodiment] (Configuration Example) (1) System Figure 1 is a diagram showing the configuration of a virtual space presentation system equipped with a virtual space generation device CSA according to the first embodiment of the present invention.

[0013] The system according to the first embodiment is configured such that a head-mounted display HD worn on the head of a user US and a user terminal TM used by the user US can be connected to a virtual space generation device CSA via a network NW.

[0014] The head-mounted display HD is used to display virtual space images transmitted from the virtual space generation device CSA. In addition to the head-mounted display, any display device capable of displaying 360-degree virtual space images, such as a multi-screen display device, may be used.

[0015] The user terminal TM is, for example, a smartphone, and is used to input information representing the user's personality traits, desired impression in the virtual space, and current state, and to transmit this information to the virtual space generation device CSA. The user terminal TM may also be a tablet device or a personal computer.

[0016] A network (NW) comprises, for example, a wide-area network centered on the Internet, and an access network for accessing this wide-area network. The access network may, but is not limited to, a public data communication network using wireless technology or a LAN (Local Area Network).

[0017] (2) Virtual Space Generator CSA The virtual space generator CSA is installed on a server computer, for example, on the cloud or on the web. Alternatively, the virtual space generator CSA may be installed on a personal computer used by a system administrator or user US.

[0018] Figures 2 and 3 are block diagrams showing examples of the hardware and software configurations of the virtual space generation device CSA, respectively.

[0019] The virtual space generation device CSA includes a control unit 1A that uses a hardware processor such as a Central Processing Unit (CPU), and a storage unit having a program storage unit 2A and a data storage unit 3A, and a communication interface (hereinafter referred to as I / F) unit 4 are connected to this control unit 1A via a bus 5.

[0020] The communication interface unit 4 uses a communication protocol defined in the network NW to send and receive various information, such as virtual space images and the user's status, between the user's head-mounted display HD and the user terminal TM.

[0021] The program storage unit 2A is configured, for example, by combining a non-volatile memory that can be written to and read at any time, such as an SSD (Solid State Drive), and a non-volatile memory such as a ROM (Read Only Memory), and stores application programs necessary to execute various controls according to the first embodiment, in addition to middleware such as an OS (Operating System). Hereafter, the OS and each application program will be collectively referred to as a program.

[0022] The data storage unit 3A combines, for example, a non-volatile memory such as an SSD that can be written to and read at any time, and a volatile memory such as RAM (Random Access Memory) as a storage medium. Its storage area is provided with a character traits storage unit 31, a target state storage unit 32, a spatial information storage unit 33, and a current state storage unit 34.

[0023] The personality traits memory unit 31 stores information representing the user's personality traits. The target state memory unit 32 stores information representing the target state of the user US. The spatial information memory unit 33 is used to store the generated virtual space image. The current state memory unit 34 stores information representing the user US's current state.

[0024] The control unit 1A includes, as processing functions necessary to carry out the first embodiment of this invention, a character trait acquisition processing unit 11A, a target state acquisition processing unit 12A, a spatial parameter determination processing unit 13A, a spatial generation processing unit 14A, a current state acquisition processing unit 15A, and a spatial parameter optimization processing unit 16A.

[0025] Each of the above-mentioned processing units 11A to 16A is implemented by having the hardware processor of the control unit 1A execute an application program stored in the program storage unit 2A. Note that some or all of the above-mentioned processing units 11A to 16A may be implemented using hardware such as LSI (Large Scale Integration) or ASIC (Application Specific Integrated Circuit).

[0026] The personality trait acquisition processing unit 11A acquires information representing the personality traits of the user US, for example from the user terminal TM, and stores the acquired information representing the personality traits in the personality trait storage unit 31.

[0027] The target state acquisition processing unit 12A acquires information representing the user's target state (hereinafter referred to as the target state) when the user US's state is changed by intervention using a virtual space, for example from the user terminal MT, and stores the acquired information representing the target state in the target state storage unit 32.

[0028] The spatial parameter determination processing unit 13A performs a process to determine the initial values ​​of spatial parameters that give the impression of being consistent with the target state, based on the acquired information representing the personality traits and target state of the user US.

[0029] The spatial generation processing unit 14A generates information representing a virtual space that gives the impression of matching the target state, based on the initial values ​​of the determined spatial parameters. The information representing the virtual space may include sound information in addition to video. The spatial generation processing unit 14A then stores the generated information representing the virtual space in the spatial information storage unit 33 and outputs it to the head-mounted display HD from the communication I / F unit 4 to present it to the user US.

[0030] The current state acquisition processing unit 15A, with the information representing the virtual space presented, acquires information representing the current state of the user US (hereinafter referred to as the current state) from, for example, the user terminal TM and stores it in the current state storage unit 34.

[0031] The spatial parameter optimization processing unit 16A optimizes the spatial parameters to bring the current state closer to the target state based on the target state, the current state, and the initial values ​​of the spatial parameters, and provides the optimized spatial parameters to the spatial generation processing unit 14A.

[0032] (Operation Example) Next, an operation example of the virtual space generation device CSA configured as described above will be explained.

[0033] Figure 4 is a flowchart showing an example of the processing procedure and processing content of the virtual space generation process executed by the control unit 1A of the virtual space generation device CSA.

[0034] For example, suppose user US inputs a request to create a virtual space on user terminal TM. The control unit 1A of the virtual space generation device CSA detects the input of the generation request in step S10 and then starts a series of processes to generate the virtual space.

[0035] (1) Personality trait acquisition The control unit 1A of the virtual space generation device CSA first acquires information representing the personality traits of user US under the personality trait acquisition processing unit 11A in step S11.

[0036] For example, the personality trait acquisition processing unit 11A acquires information from the user terminal TM that classifies a person's personality using a classification method called the BIG-Five. The BIG-Five classifies a person's personality into five factors, which are "Openness," "Conscientiousness," "Extraversion," "Agreeableness," and "Neuroticism."

[0037] In this embodiment, the personality trait acquisition processing unit 11A acquires, for example, a score for neurotic tendencies. The score is represented, for example, on an 11-point scale from "0" to "10". As an acquisition method, for example, a questionnaire sheet is distributed to the user terminal TM of the user US, and the user US's response data to the questionnaire sheet is acquired from the user terminal TM.

[0038] The personality trait acquisition processing unit 11A stores the acquired information representing the personality traits in the personality trait storage unit 31, associating it with the user ID.

[0039] The Big Five personality domains are described in detail in, for example, Reference 1 below. Reference 1: Gosling, SD, Rentfrow, PJ, & Swann, WB, Jr. (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37, 504-528.

[0040] (2) The control unit 1A of the target state acquisition virtual space generation device CSA then, in step S12, under the control of the target state acquisition processing unit 12A, acquires information from the user terminal TM that represents the target state (target state) that the user US will aim for when the user's state is changed by intervention using the virtual space.

[0041] For example, the target state acquisition processing unit 12A acquires the user US's "comfort" or "alertness level," measured by vital sensors or the like in the user terminal TM, as information representing the target state. For example, if it is "comfort," it acquires a value that scores the level of comfort at that time, with "0" representing discomfort and "1" representing comfort, within the range of "0 to 1".

[0042] The target state acquisition processing unit 12A then stores the acquired information representing the target state in the target state storage unit 32, associating it with the user ID.

[0043] The methods for measuring the above-mentioned "pleasure" or "arousal level" are described in detail in, for example, Reference 2 below, so an explanation is omitted here. Reference 2: Russell, James A. "A circumplex model of affect." Journal of personality and social psychology 39.6 (1980): 1161.

[0044] (3) Determination of spatial parameters The control unit 1A of the virtual space generation device CSA then performs a process to determine the initial values ​​of the spatial parameters under the control of the spatial parameter determination processing unit 13A in step S13.

[0045] For example, as shown in Figure 10, if we consider an intervention aimed at putting user US into a relaxed state within a room represented by the virtual space VS, spatial parameters could include the room's size, brightness, and height.

[0046] Figure 5 is a flowchart showing an example of the processing procedure and processing content of the spatial parameter determination process executed by the spatial parameter determination processing unit 13A.

[0047] The spatial parameter determination processing unit 13A first reads information representing the user US's personality traits from the personality traits storage unit 31 in step S131, and then reads information representing the target state from the target state storage unit 32 in step S132. Then, in step S133, the spatial parameter determination processing unit 13A performs a process to determine the initial values ​​of spatial parameters that give the impression of matching the target state, based on the information of the user US's personality traits and the target state that it has read.

[0048] For example, the spatial parameter determination processing unit 13A defines the impression I as I = p × w, based on the report in Non-Patent Literature 1 that the user's impression I changes depending on the spatial parameter p and the personality trait x. Here, w is a weight parameter that changes according to the value of the personality trait x. The weight parameter w may be determined based on research, for example, as described in Non-Patent Literature 1.

[0049] For example, let the spatial parameter p be the size of a room represented by a virtual space [m²]. 2 Assuming that the weight parameter w is w = 0.4 obtained by investigating the relationship between neurotic tendencies, room size, and comfort in a large number of people, the spatial parameter p for which I = y is p = y / w = 0.8 / 0.4 = 2[m 2 This is determined to be the case.

[0050] (4) Generation of virtual space The control unit 1A of the virtual space generation device CSA then generates information representing a virtual space that has the impression of matching the target state, under the control of the space generation processing unit 14A, based on the initial values ​​of the space parameters determined by the space parameter determination processing unit 13A.

[0051] For example, when representing the inside of a room using a virtual space VS as shown in Figure 10, the size of the room is represented by the spatial parameter p, which is 2 [m²]. 2 This generates video of the virtual space set to [ ].

[0052] The spatial generation processing unit 14A stores the generated virtual space information in the spatial information storage unit 33 and transmits it from the communication interface unit 4A to the user's head-mounted display HD. As a result, the virtual space image is displayed on the head-mounted display HD.

[0053] (5) Acquisition of current state With the virtual space image presented to the user US, the control unit 1A of the virtual space generation device CSA acquires information representing the current state of the user US under the control of the current state acquisition processing unit 15A in step S15.

[0054] For example, the current status acquisition processing unit 15A reads pre-prepared questionnaire information from the data storage unit 3A and transmits the read questionnaire information to the user terminal TM of the user US via the communication interface unit 4A. The questionnaire information may be in the form of text data, audio data, or both, and may also be transmitted to the head-mounted display HD.

[0055] When a user US inputs and returns the above-mentioned questionnaire information, for example, on a user terminal TM, the current state acquisition processing unit 15A receives the above-mentioned answer information via the communication I / F unit 4A and stores the received answer information in the current state storage unit 34, as information representing the current state of the user US, and associating it with the user ID.

[0056] (6) Spatial parameter optimization The control unit 1A of the virtual space generation device CSA then performs the following process in step S16 to optimize the spatial parameters in order to bring the current state of the user US closer to the target state, under the control of the spatial parameter optimization processing unit 16A.

[0057] Figure 6 is a flowchart showing an example of the processing procedure and processing content of the spatial parameter optimization process executed by the spatial parameter optimization processing unit 16A.

[0058] In other words, the spatial parameter optimization processing unit 16A first reads information representing the user US's target state from the target state storage unit 32 in step S161, reads the acquired information representing the current state from the current state storage unit 34 in step S162, and then in step S163 obtains the initial values ​​of the spatial parameters previously determined by the spatial parameter determination processing unit 13A.

[0059] In step S164, the spatial parameter optimization processing unit 16A then performs calculations to optimize the spatial parameters based on the target state, the current state, and the initial values ​​of the spatial parameters.

[0060] For example, the spatial parameter optimization processing unit 16A processes the target state y and the current state y^ ｉ The relationship between the difference and the optimal value p' of the spatial parameter is given by y + (y - y^ ｉ It is defined as ) = w × p', where w is the weight parameter.

[0061] Then, the spatial parameter optimization processing unit 16A determines the optimal value p' of the spatial parameter based on the above equation as p' = y + (y - y^ ｉ Calculated using ) / w.

[0062] For example, if the target state y is 0.8 and the current state y^ ｉ If we set w to 0.6 and w to 0.4, the optimal value of the space parameter p' is p' = y + (y - y^ ｉ ) / w =0.8+(0.8-0.6) / 0.4 =2.5[m2 is calculated.

[0063] The spatial parameter optimization processing unit 16A outputs the optimal value p' of the calculated spatial parameter to the space generation processing unit 14A in step S165. As a result, the space generation processing unit 14A generates a virtual space image in which the optimal value p' of the spatial parameter is reflected, and transmits it from the communication I / F unit 4A to the head-mounted display HD. As a result, a virtual space in which the spatial parameter is optimized is displayed on the head-mounted display HD.

[0064] (7) End of space generation When the control unit 1A of the virtual space generation device CSA finishes the optimization processing of the spatial parameter, it makes an end determination of the space generation processing in step S17. In this state, for example, when the user US performs an operation to end viewing the virtual space, the control unit 1A returns to the standby state.

[0065] On the other hand, if the operation to end viewing the virtual space is not performed, the control unit 1A returns to step S14 and repeatedly executes the processing from step S14 to step S17. That is, the control unit 1A repeatedly executes the generation of the virtual space, the acquisition of the current state of the user US, and the optimization processing of the spatial parameter based on the acquired current state.

[0066] Therefore, the spatial parameter is sequentially optimized according to the change in the current state of the user US, and thereby it becomes possible to make the impression of the virtual space closer to the target state of the user US himself / herself.

[0067] In the above description, the spatial parameter is repeatedly optimized according to the change in the current state of the user US. However, if the virtual space state can be optimized to the target state by one optimization process, the repeated optimization process does not necessarily have to be performed.

[0068] (Effects) As described above, in the first embodiment, first, information representing the personality traits of the user US and the target state that the user US seeks in the virtual space is acquired. Based on the acquired information, initial values ​​of spatial parameters that give the impression of matching the target state are determined, and a virtual space reflecting the set initial values ​​of the spatial parameters is presented to the user US. Next, information representing the current state of the user US after the presentation of the virtual space is acquired, the acquired current state is optimized to bring it closer to the target state, and a virtual space reflecting the optimized spatial parameters is generated and presented to the user US.

[0069] Therefore, even if the impression received from the virtual space, which reflects the initial values ​​of the spatial parameters, deviates from the user's desired target state, the user's current state is acquired, and the spatial parameters are optimized so that this current state approaches the target state. Virtual space information reflecting the optimized spatial parameters is then generated. As a result, a virtual space reflecting the user's current state can be generated and presented to the user.

[0070] [Second Embodiment] A second embodiment of this invention proposes a method for optimizing spatial parameters when there are multiple spatial parameters.

[0071] In other words, when there are multiple spatial parameters, they influence each other, making it difficult to determine the optimal space with a single correction. Therefore, we employ a method that optimizes spatial parameters while reducing the burden on the user when performing multiple corrections. One example of an optimization method is Bayesian optimization. Bayesian optimization is a method that obtains the optimal value by collecting a small amount of data, and employs a mechanism that repeatedly learns to suggest parameters to explore by taking a score for a certain parameter as input.

[0072] (Configuration Example) Figure 7 is a block diagram showing an example of the software configuration of a virtual space generation device CSB according to the second embodiment of the present invention. In Figure 7, the same functional parts as in Figure 3 are denoted by the same reference numerals, and detailed explanations are omitted.

[0073] In addition to the personality trait acquisition processing unit 11A, the target state acquisition processing unit 12A, and the current state acquisition processing unit 15A, the control unit 1B includes a spatial parameter determination processing unit 13B, a spatial generation processing unit 14B, and a spatial parameter optimization processing unit 16B as processing functions necessary for implementing the second embodiment.

[0074] Furthermore, the above-mentioned processing units 13B to 16B, together with processing units 11A, 12A, and 15A, are realized by causing the hardware processor of the control unit 1B to execute the application program stored in the program storage unit 2B.

[0075] The spatial parameter determination processing unit 13B determines the nth spatial parameter p presented i times that has an impression I that matches the target state y, based on the personality trait x and the target state y. ｍｉ Determine the initial value of m. Note that m is a corresponding integer between 1 and N.

[0076] The spatial generation processing unit 14B processes the spatial parameter p determined by the spatial parameter determination processing unit 13B. ｍｉ Based on the initial values ​​and the spatial parameters optimized by the spatial parameter optimization processing unit 16B described later, a virtual space reflecting these spatial parameters is generated.

[0077] The spatial parameter optimization processing unit 16B comprises an evaluation function application processing unit 161, a model learning processing unit 162, and a spatial parameter candidate selection processing unit 163. In addition, corresponding to these processing functions, the data storage unit 3B is provided with an answer result storage unit 35 for storing the evaluation function application results and a learning model storage unit 36 ​​for storing the learning model used for model learning.

[0078] The evaluation function application processing unit 161 evaluates the target state y and the current state y^ ｉ The evaluation function value f for optimization based on this. ｉ The evaluation function application processing unit 161 then calculates the above evaluation function value f. ｉ The i-th spatial parameter p determined by the spatial parameter determination processing unit 13B described above ｍｉ The corresponding result is stored in the answer result storage unit 35.

[0079] The model learning processing unit 162 retrieves the i-th spatial parameter p from the answer result storage unit 35. ｍｉ And the evaluation function f ｉ The above spatial parameter p is read and read. ｍｉ And the evaluation function f ｉ Based on this, we calculate candidate spatial parameters to be reflected in the virtual space presented next. A Bayesian optimization learning model is used for this calculation process.

[0080] The spatial parameter candidate selection processing unit 163 selects optimized spatial parameters from the spatial parameter candidates calculated by the model learning processing unit 162 and provides the selected spatial parameters to the spatial generation processing unit 14B.

[0081] (Example of operation) (1) Determination of spatial parameters The control unit 1B of the virtual space generation device CSB determines the initial values ​​of the spatial parameters under the control of the spatial parameter determination processing unit 13B as follows.

[0082] Figure 8 is a flowchart showing an example of the processing procedure and processing content of the spatial parameter determination process executed by the spatial parameter determination processing unit 13B. In Figure 8, the same parts as in Figure 5 are denoted by the same reference numerals for explanation.

[0083] In other words, the spatial parameter determination processing unit 13B first reads the user US's personality traits x from the personality trait storage unit 31 in step S131, and reads the user US's target state y in step S132. Then, in step S134, the spatial parameter determination processing unit 13B, based on the read personality traits x and target state y, selects the nth spatial parameter p presented i times that has an impression I that matches the target state y. ｍｉ To decide.

[0084] For example, the spatial parameter determination processing unit 13B determines that, as described in Non-Patent Document 1, the user's impression I is a spatial parameter p ｍｉ Based on reports that it changes depending on personality trait x, the impression I is given by I = Σ N m-1　 lol ｍ ×p ｍｉThis is defined as follows: Here, w ｍ This is a weight parameter that changes according to the value of personality trait x. Note that the weight parameter w ｍ This may be determined based on research, for example, as described in Non-Patent Document 1.

[0085] For example, now the spatial parameter p １１ The size of the room represented in virtual space [m²] 2 ], spatial parameter p ２１ The ceiling height [m], spatial parameter p ３１ Let be the brightness of the light [lumens]. Also, the weight parameter w ｍ This is a value obtained by investigating the relationship between neurotic tendencies and room size, ceiling height, and lighting brightness in a large number of people, for example, w １ = 0.1, w ２ = 0.2, w ３ Let = 0.01. Under these conditions, the spatial parameter p such that I = y １１ , p ２１ , p ３１ Since y = 0.8, we have 0.8 = 0.1p. １１ +0.2p ２１ +0.01p ３１ It is represented as follows.

[0086] The spatial parameter p expressed by this conditional expression １１ , p ２１ , p ３１ There are infinitely many combinations of the above spatial parameter p. １１ , p ２１ , p ３１ The value of is randomly determined within the range that satisfies the above condition. Alternatively, the spatial parameter p １１ , p ２１ , p ３１ It may be determined using some constraints for each. Incidentally, in this embodiment (p １１ , p ２１ , p ３１ ) = (2, 1, 400) is determined.

[0087] (2) Generation of virtual space Virtual space generation device CS The control unit 1B then generates information representing a virtual space that has the impression of matching the target state, under the control of the space generation processing unit 14B in step S14, as shown in Figure 4, based on the initial values ​​of the space parameters determined by the space parameter determination processing unit 13B.

[0088] For example, when representing the inside of a room using a virtual space VS as shown in Figure 10, the space generation processing unit 14B uses the above-mentioned space parameter p to determine the state of the room. １１ , p ２１ , p ３１ Area represented by = 2 [m 2 It generates a virtual space image set to 1 m in height and 400 lumens in brightness.

[0089] The spatial generation processing unit 14B stores the generated virtual space information in the spatial information storage unit 33 and transmits it from the communication I / F unit 4A to the user's head-mounted display HD. As a result, the head-mounted display HD displays a virtual space image that reflects the width, height, and brightness as described above.

[0090] (3) Current State Acquisition The control unit 1B of the virtual space generation device CSB, in the same manner as in the first embodiment, under the control of the current state acquisition processing unit 15A, in step S15, acquires information y^ representing the i-th current state of user US. ｉ The current state is obtained. As described in the first embodiment, a questionnaire survey is used as a method for obtaining the current state, but other methods may also be used.

[0091] (4) Optimization of spatial parameters The current state y^ of the above user US ｉ Once acquired, the control unit 1B of the virtual space generation device CSB performs spatial parameter optimization processing under the control of the spatial parameter optimization processing unit 16B as follows.

[0092] Figure 9 is a flowchart showing an example of the processing procedure and content of the spatial parameter optimization process performed by the spatial parameter optimization processing unit 16B. In this figure, the same parts as in Figure 6 are denoted by the same reference numerals for explanation.

[0093] In other words, the spatial parameter optimization processing unit 16B first reads the target state y of the user US from the target state storage unit 32 in step S161, and then reads the current state y^ from the current state storage unit 34 in step S162. ｉ The spatial parameter optimization processing unit 16B then reads the target state y read by the evaluation function application processing unit 161 and the current state y^ ｉ Based on this, the evaluation function value for optimizing the spatial parameters is calculated (step S166). Note that the evaluation function can be arbitrarily selected as long as it is the function that is maximized in the target state y.

[0094] In this embodiment, the evaluation function f for the ith time ｉ to f ｉ = |y - y^ ｉ | is defined as follows. Then, the evaluation function application processing unit 161 calculates the evaluation function value f based on the above formula. ｉ This calculates the result, for example, if the target state y is 0.8 and the current state y^ ｉ If it is 0.2, then the evaluation function f ｉ This is calculated to be 0.6.

[0095] The evaluation function application processing unit 161 applies the calculated evaluation function f ｉ The spatial parameter p, which was previously determined by the spatial parameter determination processing unit 13B. ｍｉ At the same time, the result of the i-th response is stored in the response result storage unit 35 (step S167). Figure 11 shows an example of a response stored in the response result storage unit 35.

[0096] The spatial parameter optimization processing unit 16B then uses the model learning processing unit 162 to process the spatial parameters p stored in the answer result storage unit 35. ｍｉ And the evaluation function f ｉ Based on this, candidate spatial parameters to be reflected in the virtual space presented next are calculated (step S168).

[0097] One method for calculating the spatial parameters mentioned above is Bayesian optimization using a learning model. Bayesian optimization is an optimization method that obtains the optimal value by collecting a small amount of data. It takes a score for a certain parameter as input and repeatedly learns to suggest parameters to explore. Hyperparameters for Bayesian optimization include acquisition functions, but any function that is empirically suitable can be used.

[0098] Bayesian optimization is explained in detail in, for example, reference 3 below, so a detailed explanation will be omitted here. Reference 3: Shahriari, Bobak, et al. "Taking the human out of the loop: A review of Bayesian optimization." Proceedings of the IEEE 104.1 (2015): 148-175. Gosling, SD, Rentfrow, PJ, & Swann, WB, Jr. (2003).

[0099] The model learning processing unit 162, for example, in the first optimization process i=1, processes the spatial parameter p １１ , p ２１ , p ３１ Since = (2, 1, 400), these spatial parameters p １１ , p ２１ , p ３１ And the evaluation function f ｉ The relationship with = 0.6 is trained in a learning model employing Bayesian optimization. Subsequently, steps S14 to S16 shown in Figure 4 are repeated in the same manner, and the spatial parameter p for each of the i = 2nd, 3rd, ... times is trained. ｍｉ and the corresponding evaluation function f ｉ The learning model is repeatedly trained on the relationship between these two factors.

[0100] Figure 12 shows the spatial parameters p stored in the answer result storage unit 35 after the optimization process using Bayesian optimization has been repeatedly performed as described above. ｍｉ and evaluation function f ｉ This figure shows an example of the relationship.

[0101] While repeatedly executing the above i-th process, the spatial parameter optimization processing unit 16B determines in step S169 whether the current state y^ ｉ matches the target state y. When the current state y^ ｉ matches the target state y, the spatial parameter candidate at this time is selected, and the selected spatial parameter is output to the space generation processing unit 14B in step S165.

[0102] For example, when the result shown in FIG. 12 is obtained as a result of the i-th repeated process, the spatial parameter candidate selection processing unit 163 determines that the current state y^ ｉ matches the target state y = 0.8, that is, when the evaluation function value f ｉ = 0 (i = 6th time), the corresponding spatial parameter p １６ , p ２６ , p ３６ , which are "0.95", "2.09", and "400", is selected. Then, the selected spatial parameters p １６ , p ２６ , p ３６ are output to the space generation processing unit 14B.

[0103] As a result, in the space generation processing unit 14B, a virtual space in which the above spatial parameters p １６ , p ２６ , p ３６ are reflected, that is, a virtual space with a width of 0.95 [m ２ , a ceiling height of 2.09 [m], and a lighting brightness of 400 [lumens] is generated. Then, the generated virtual space video is transmitted from the communication I / F unit 4A to the head-mounted display HD of the user US and displayed.

[0104] The end condition of the above i-th repeated process may be when the difference between the current state y^ ｉ and the target state y, that is, the evaluation function value f ｉ becomes less than or equal to a preset value, or when the number of repetitions reaches a predetermined upper limit value.

[0105] (Effects) As described above, in the second embodiment, when there are multiple spatial parameters, a series of processes from the generation of a virtual space to the acquisition of the current state and the optimization of spatial parameters are repeatedly executed using a learning model that employs Bayesian optimization, under the control of the spatial parameter optimization processing unit 16B. Then, when termination conditions such as the evaluation function value falling below a threshold are met, candidate spatial parameters are selected, and virtual space information reflecting the selected spatial parameters is presented to the user US.

[0106] Therefore, even when there are multiple spatial parameters, the optimal spatial parameters are selected in a short time, and that virtual space information is presented to the user (US). As a result, it becomes possible to quickly generate and present an optimal virtual space that reflects the target state desired by the user (US) and is defined by multiple spatial parameters.

[0107] [Other Embodiments] (1) In the first and second embodiments, the processing functions of the virtual space generation devices CSA and CSB were described as being provided on a server computer on the Web or in the cloud. However, the processing functions of the virtual space generation devices CSA and CSB may be provided on a personal computer used by the system administrator or user. Alternatively, the processing functions of the virtual space generation devices CSA and CSB may be divided and located on multiple server computers or personal computers.

[0108] (2) In addition, the configuration of the processing functions of the virtual space generation device, its processing procedures and contents, in particular the optimization algorithm for spatial parameters, the types of virtual spaces, etc., can be modified in various ways without departing from the spirit of this invention.

[0109] Although embodiments of this invention have been described in detail above, the above description is merely illustrative in all respects. It goes without saying that various improvements and modifications can be made without departing from the scope of this invention. In other words, when implementing this invention, specific configurations may be adopted as appropriate depending on the embodiment.

[0110] In short, this invention is not limited to the embodiments described above, and the components can be modified and implemented in practice without departing from the gist of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the embodiments described above. For example, some components may be deleted from all the components shown in the embodiments. Moreover, components from different embodiments may be appropriately combined.

[0111] CSA, CSB...Virtual Space Generation Device US...User HD...Head-Mounted Display TM...User Terminal 1A, 1B...Control Unit 2A, 2B...Program Storage Unit 3A, 3B...Data Storage Unit 4A, 4B...Communication I / F Unit 5A...Bus 11...Personality Trait Acquisition Processing Unit 12A...Target State Acquisition Processing Unit 13A, 13B...Spatial Parameter Determination Processing Unit 14A, 14B...Space Generation Processing Unit 15A...Current State Acquisition Processing Unit 16A, 16B...Spatial Parameter Optimization Processing Unit 161...Evaluation Function Application Processing Unit 162...Model Learning Processing Unit 163...Spatial Parameter Candidate Selection Processing Unit 31...Personality Trait Storage Unit 32...Target State Storage Unit 33...Spatial Information Storage Unit 34...Current State Storage Unit 35...Answer Result Storage Unit 36...Learning Model Storage Unit

Claims

1. A virtual space generation device comprising: a first processing unit that acquires information representing the user's personality traits and information representing the user's target state in the virtual space; a second processing unit that determines spatial parameters corresponding to an impression consistent with the target state based on the personality traits and the target state; a third processing unit that acquires information representing the user's current state when the virtual space information reflecting the spatial parameters is presented to the user; and a fourth processing unit that optimizes the spatial parameters to bring the current state closer to the target state.

2. The virtual space generation device according to claim 1, wherein the second processing unit determines a plurality of spatial parameters corresponding to an impression that matches the target state, and the fourth processing unit repeatedly performs a series of processes from presenting the virtual space information reflecting the spatial parameters to acquiring information representing the current state and optimizing the plurality of spatial parameters a plurality of times until the current state approaches the target state within a predetermined range of difference.

3. A virtual space generation method executed by an information processing device, comprising: a process of acquiring information representing the user's personality traits and information representing the user's own target state that the user seeks in the virtual space; a process of determining a first spatial parameter corresponding to an impression consistent with the target state based on the personality traits and the target state; a process of generating first virtual space information reflecting the first spatial parameter and presenting it to the user; a process of acquiring information representing the user's current state while the first virtual space information has been presented to the user; a process of optimizing the first spatial parameter to bring the current state closer to the target state and generating a second spatial parameter; and a process of generating second virtual space information reflecting the second spatial parameter and presenting it to the user.

4. A program that causes a processor in a virtual space generation device to execute at least one of the processes performed by the first to fourth processing units in the virtual space generation device described in claim 1 or claim 2.

Images