Real estate image adjustment system, real estate image adjustment apparatus, real estate image adjustment method, and real estate image adjustment program

Abstract

To provide a real estate image adjustment system, a real estate image adjustment apparatus, a real estate image adjustment method, and a real estate image adjustment program which allow a user to easily adjust an image quality and composition of real estate image data in a short time even if the user is not familiar with image adjustment.SOLUTION: A real estate image adjustment system according to the present invention includes a user terminal and a server. The server estimates an image quality and composition of real estate image data transmitted from the user terminal (at step S202), and proposes proper parameter values while recommending retouch item data necessary for adjusting a recommended image (at step S203). The server automatically adjusts the image quality and composition of the real estate image data based on the parameter values to those of the recommended image (at step S204). The user terminal displays, on a display unit, the retouch item data and the recommended image transmitted from the server. Thus, the user can compare the recommended image with the input original real estate image data and can edit it to a preferable image.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to a real estate image adjustment system, a real estate image adjustment device, a real estate image adjustment method, and a real estate image adjustment program, and particularly to a real estate image adjustment system, a real estate image adjustment device, a real estate image adjustment method, and a real estate image adjustment method capable of adjusting the image quality and composition of real estate image data.

Background Art

[0002] Conventionally, as a tool for editing image data of real estate such as buildings (hereinafter referred to as "real estate image data"), software capable of adjusting image quality and composition (hereinafter referred to as "image editing software") has been widely known (see Non-Patent Document 1). According to such image editing software, by using this, it is possible for a user to obtain real estate image data similar to that obtained when a professional cameraman images real estate.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in recent years, with the improvement of the image quality of imaging devices (for example, cameras), image editing software with more advanced image adjustment functions has become mainstream. Generally, image adjustment using this type of image editing software is performed by individually adjusting adjustment items (hereinafter referred to as "retouch items") that consist of white balance (color temperature, color cast), color tone (exposure, contrast, brightness, etc.), color hue (sharpness, saturation, etc.), effects (edge strength, etc.), and so on, one by one.

[0005] That is, in conventional image editing software, proficient users such as professional photo retouchers can fully utilize its functions. However, beginners who are not familiar with the functions often have difficulty understanding which retouch items to adjust. In such cases, not only does it take time to adjust the image quality, but in some cases, there are problems such as being unable to obtain satisfactory real estate image data with high image quality.

[0006] Also, generally, since images vary from person to person in terms of preferences and ways of thinking, there is no clear criterion for determining how to adjust them to an appropriate quality. Therefore, even if there is image editing software as described above, especially for those who are not proficient in image adjustment (beginners), there is a problem that they cannot fully utilize it and cannot obtain high-quality images that meet their needs.

[0007] The present invention has been made to solve such problems, and provides a real estate image adjustment system, a real estate image adjustment device, a real estate image adjustment method, and a real estate image adjustment program that enable even users who are not accustomed to image quality adjustment to easily and quickly adjust the image quality of real estate image data.

Means for Solving the Problems

[0008] According to the real estate image adjustment system of the present invention, the above problem is solved by an image adjustment system including a user terminal having an input means capable of inputting real estate image data obtained by imaging a real estate, and a server for adjusting the image quality and composition of the real estate image data transmitted from the user terminal. The server includes a storage means for storing a plurality of retouch item data indicating adjustment items for the image quality and composition of the real estate image data, an image estimation means for estimating the image quality and composition of the real estate image data transmitted from the user terminal using a learning model machine-learned with past real estate image data as teacher data, a retouch item data recommendation means for selecting and recommending, from the storage means, the retouch item data necessary for adjusting the image quality and composition of the real estate image data estimated by the image estimation means to a recommended image based on the learning model, an image adjustment means for adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by the retouch item data recommendation means, and a transmission means for transmitting the adjusted real estate image data adjusted by the image adjustment means to the user terminal together with the recommended retouch item data. The user terminal includes a display means for displaying the adjusted real estate image data and the recommended retouch item data transmitted by the transmission means. Here, the "real estate image data" means image data obtained by imaging the interior and exterior views, furniture, land, and surrounding environment of a real estate property (the "real estate").

[0009] In the invention related to the real estate image adjustment system, it is preferable that the input means is capable of adjusting the recommended retouch item data transmitted from the transmission means, and the image adjustment means adjusts the image quality and composition of the real estate image data based on the adjusted real estate image data and the adjusted retouch item data input by the input means.

[0010] Furthermore, in the invention related to the real estate image adjustment system, the server further includes a recommended image value calculation means for calculating a recommended image value corresponding to the recommended image for each of the retouch item data, the transmission means transmits data indicating the recommended image value for each of the retouch item data to the user terminal, and the display means preferably displays the data indicating the recommended image value transmitted by the transmission means together with the retouch item data.

[0011] According to the real estate image adjustment device according to the present invention, the above problems can also be solved by including: an input means capable of inputting the real estate image data; a storage means for storing a plurality of retouch item data indicating adjustment items for the image quality and composition of the real estate image data; an image estimation means for estimating the image quality and composition of the real estate image data input by the input means using a learning model machine-learned with the past real estate image data as teacher data; a retouch item data recommendation means for selecting and recommending, from the storage means, the retouch item data necessary for adjusting the image quality and composition of the real estate image data to a recommended image based on the learning model; an image adjustment means for adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by the retouch item data recommendation means; and a display means for displaying the adjusted real estate image data adjusted by the image adjustment means together with the recommended retouch item data.

[0012] Further, according to the real estate image adjustment method of the present invention, the above problem can also be solved by a real estate image adjustment method capable of adjusting the image quality and composition of real estate image data obtained by imaging real estate, including: an input step of inputting the real estate image data; an image estimation step of estimating the image quality and composition of the real estate image data input by performing the input step using a learning model machine-learned with the past real estate image data as teacher data; a retouch item data recommendation step of selecting and recommending, from storage means, retouch item data necessary to adjust the image quality of the real estate image data estimated by performing the image estimation step to a recommended image based on the learning model; an image adjustment step of adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by performing the retouch item data recommendation step; and a display step of displaying, on a display means, the adjusted real estate image data adjusted by performing the image adjustment step together with the recommended retouch item data.

[0013] Furthermore, according to the real estate image adjustment program of the present invention, the above problem can also be solved by an image adjustment program capable of adjusting the image quality and composition of real estate image data obtained by imaging real estate, including: an image estimation step of estimating the image quality and composition of the real estate image data input from an input means using a learning model machine-learned with the past real estate image data as teacher data; a retouch item data recommendation step of selecting and recommending, from storage means, retouch item data necessary to adjust the image quality and composition of the real estate image data estimated by performing the image estimation step to a recommended image based on the learning model; an image adjustment step of adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by performing the retouch item data recommendation step; and a display step of displaying, on a display means, the adjusted real estate image data adjusted by performing the image adjustment step together with the recommended retouch item data, and causing a computer to execute these steps.

Advantages of the Invention

[0014] As described above, according to the real estate image adjustment system, real estate image adjustment device, real estate image adjustment method, and real estate image adjustment program according to the present invention, when adjusting an image, guidance for appropriately performing the adjustment is presented. Therefore, even a user who is not familiar with image adjustment can easily obtain a high-quality real estate image in a short time (a user who is not familiar with it can obtain the necessary image just by uploading the image. Also, when a user who is not familiar with it wants to change a real estate image, it can be easily corrected using the suggestion function by artificial intelligence (AI)). In addition, in the real estate image adjustment system, real estate image adjustment device, real estate image adjustment method, and real estate image adjustment program according to the present invention, since it is possible to provide a qualified index for obtaining an image of a certain quality, it is also possible to guarantee that the image meets the necessary criteria.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0016] Hereinafter, the real estate image adjustment system of the present invention will be described with reference to the drawings based on a preferred embodiment thereof. FIG. 1 is a schematic diagram showing an overview of the real estate image adjustment system 1 according to this embodiment, FIG. 2 is a block diagram of the user terminal 10, FIG. 3 is a block diagram of the server 20, FIG. 4 is a flowchart showing the content of control processing in the user terminal 10, and FIG. 5 is a flowchart showing the content of control processing in the server 20.

[0017] <Overall Configuration of Real Estate Image Adjustment System 1> As shown in FIG. 1, the real estate image adjustment system 1 according to this embodiment is a system for performing image adjustment of real estate image data (see FIG. 6), and is configured to include a user terminal 10 and a server 20. The user terminal 10 and the server 20 are connected to each other via a public communication line 2. Note that the real estate image adjustment system 1, the real estate image data, the user terminal 10, and the server 20 respectively correspond to the "image adjustment system", "real estate image data", "user terminal", and "server" described in the claims.

[0018] Such real estate image data is image data obtained by imaging real estate such as the exterior and interior of a building, land, and the surrounding environment, and can be created using known file formats such as TIFF (Tagged Image File Format), JPEG (Joint Photographic Experts Group), PNG (Portable Network Graphics), and Bitmap.

[0019] Although it will be described in detail later, the adjustment of the image quality and composition of the real estate image data (hereinafter simply referred to as "image adjustment") in this embodiment is, for example, (1) Transmitting the real estate image data from the user terminal 10 to the server 20, (2) On the server 20 side, estimating the image quality and composition of the real estate image data, and selecting and recommending retouch item data (hereinafter referred to as "retouch item data Ri", see FIG. 8) necessary for adjusting to a recommended image. (3) On the server 20 side, calculate the parameter values of each retouch item data Ri (hereinafter, also referred to as "recommended retouch item data Ri") recommended in (2) above. (4) To obtain a recommended image, adjust the image quality and composition of the real estate image data based on each data obtained in (2) and (3) above. (5) Based on the retouch item data Ri and the like transmitted from the server 20, perform further adjustment operations on the user terminal 10 side. It is performed in the following procedure. As a result, it becomes possible to obtain real estate image data according to the user's preference based on the recommended image.

[0020] <User Terminal 10> As shown in FIGS. 1 and 2, the user terminal 10 according to the present embodiment includes a central processing unit 11, a storage unit 12, a display unit 13, an input unit 14, and a communication unit 15. The user terminal 10 can be configured using, for example, a known personal computer 10A (Personal Computer) or a mobile terminal 10B (in this embodiment, a "smartphone"). Note that the display unit 13 and the input unit 14 correspond to the "display means" and the "input means" described in the claims.

[0021] (Central Processing Unit 11) The central processing unit 11 reads various programs stored in the storage unit 12, performs predetermined arithmetic processing, and executes general control to display a predetermined image on the display unit 13. Although details will be described later, the central processing unit 11 according to the present embodiment follows the image editing program 12A stored in the storage unit 12. · Control to transmit the input real estate image data (hereinafter referred to as "input image data Ii", see FIGS. 6 and 7) to the server 20 (real estate image data transmission process). · Control to display the real estate image data with adjusted image (hereinafter referred to as "image automatic adjustment data Ia", see FIG. 7) transmitted from the server 20 on the display unit 13 (image comparison display process ("step S103" in FIG. 4)). · Control to display retouch item data Ri (see FIG. 8) etc. transmitted from server 20 on display unit 13 (image editing display process ("Step S106" in FIG. 4)), · When there is further image adjustment for the image automatic adjustment data Ia displayed on display unit 13, control to change the image of the real estate image data as needed and display this on display unit 13 (image data change display process ("Step S109" in FIG. 4)), and so on.

[0022] (Memory unit 12) Memory unit 12 consists of a semiconductor memory such as a ROM (Read Only Memory) or a RAM (Random Access Memory), and an image editing program 12A etc. are stored in its storage area.

[0023] Image editing program 12A is a program for adjusting the image of real estate image data. Specifically, in image editing program 12A, instruction procedures etc. for causing central processing unit 11 to execute the above-described real estate image data transmission process, image comparison display process ("Step S103" in FIG. 4), image editing display process ("Step S106" in FIG. 4), image data change display process ("Step S109" in FIG. 4), etc. are programmed. Also, image editing program 12A is configured to be interlocked with image adjustment program 22A of server 20 described later. Thereby, when central processing unit 11 receives predetermined data, command signals, etc. from server 20, it can continuously execute related processes (see FIGS. 7 and 8).

[0024] Retouch item data Ri used when adjusting the image of real estate image data is stored in image editing program 12A. Retouch item data Ri is data that is displayed on the display unit 13 together with the real estate image data when the user adjusts the image of the real estate image data. For example, as shown in FIG. 8, it can be configured to include a retouch item name (e.g., "White Balance (Color Temperature)") and a horizontal slider having a knob icon Pi. With this configuration, by moving the knob icon Pi left and right, the user can surely grasp that the parameter value of the retouch item data Ri increases or decreases. In this case, it is preferable to display the current parameter value of the retouch item data Ri adjacent to the horizontal slider (the "parameter display unit Pv" in FIG. 8). Thereby, the user can grasp at a glance the parameter value of the current retouch item data Ri.

[0025] Examples of such retouch items include · Items such as "Color Temperature" and "Color Cast" used when adjusting "White Balance", · Items such as "Exposure" and "Contrast" used when adjusting "Color Tone", · Items such as "Clarity" and "Chroma" used when adjusting "Color Hue", · Items such as "Edge Strength" used when adjusting "Effect" (see FIG. 8). Note that the retouch items are not limited to items related to the image quality adjustment of the real estate image data, and it is also possible to add items other than image quality adjustment, for example, items for changing the composition such as the inclination and size of the image.

[0026] In the present embodiment, the retouch item data Ri programmed in the image editing program 12A is configured to be stored in the retouch item database 22D of the server 20 described later. Thereby, when the central processing unit 11 receives the retouch item data Ri from the server 20 side, it can be directly displayed on the display unit 13 or the like without performing data conversion or the like.

[0027] (Display unit 13 · Input unit 14 · Communication unit 15) As shown in FIGS. 1 and 2, the display unit 13 is, for example, a liquid crystal display (LCD: Liquid Crystal Display), and is a device that displays a predetermined image based on a command from the central processing unit 11 (see FIGS. 6 to 8). The input unit 14 is, for example, composed of a known keyboard, mouse, and touch panel, and is used not only for adding, changing, deleting, etc. various data stored in the storage unit 12, but also for importing real estate image data into the storage unit 12. The communication unit 15 is an interface capable of communicating with communication devices such as the server 20 via the public communication line 2.

[0028] <Server 20> As shown in FIGS. 1 and 3, the server 20 according to the present embodiment includes a central processing unit 21, a storage unit 22, a display unit 23, an input unit 24, and a communication unit 25. Similar to the user terminal 10, the server 20 can also be configured using, for example, a known personal computer (Personal Computer). Note that the central processing unit 21, the storage unit 22, and the communication unit 25 respectively correspond to the "image estimation means", "retouch item data recommendation means", "image adjustment means", and "recommended image value calculation means", the "storage means", and the "transmission means" described in the claims.

[0029] (Central Processing Unit 21) As shown in FIG. 3, the central processing unit 21 reads various programs stored in the storage unit 22, performs predetermined arithmetic processing, and executes general control to display the arithmetic result on the display unit 23.

[0030] Although details will be described later, the central processing unit 21 according to the present embodiment, in addition to performing the above-described general control, according to the image adjustment program 22A described later, · Controls to generate a learning model for performing machine learning (Machine Learning) using past real estate image data (hereinafter referred to as "past real estate image data") as teacher data and extracting its features (learning model generation process), · Using the machine-learned learning model, extract the feature amounts of the input image data Ii (see FIGS. 6 and 7), and perform control to estimate the image quality and composition (image estimation process ("step S202" in FIG. 5)), · Control to select and recommend the retouch item data Ri (see FIG. 8) for making the image quality and composition of the input image data Ii into a recommended image based on the learning model (recommended image value calculation process ("step S203" in FIG. 5)), · Control to obtain a value for making a recommended image (hereinafter referred to as "recommended image value Iv1", see FIG. 8) for each retouch item data Ri (recommended image value calculation process ("step S203" in FIG. 5)), · Control to automatically adjust the image quality and composition of the input image data Ii using the recommended image value Iv1 to generate the image automatic adjustment data Ia (see FIG. 7) (image automatic adjustment data generation process ("step S204" in FIG. 5)), · Control to transmit the retouch item data Ri, the recommended image value Iv1, and the image automatic adjustment data Ia to the user terminal 10 (various data transmission process ("step S205" in FIG. 5)), and the like are performed. Note that the above-described image adjustment program 22A, the recommended image, and the recommended image value Iv1 respectively correspond to the "real estate image adjustment program", the "recommended image", and the "recommended image value" described in the claims.

[0031] In the present embodiment, when performing the above-described image automatic adjustment data generation process ("step S204" in FIG. 5), elements other than the image quality, for example, defects in the composition such as the inclination and size of the image are also detected and automatically adjusted. As a result, it is possible to ensure that the finally adjusted image is not only of high quality in terms of image quality but also has a visually attractive and professional composition.

[0032] Here, the above-described learning model generation process will be described. The other processes described above will be described later. In this embodiment, appropriate-quality and well-composed past real estate image data is used as teacher data for machine learning, and a learning model is configured to be generated thereby. Examples of such past real estate image data include real estate image data captured by professional cameramen and real estate image data adjusted by professional photo retouchers. In this case, it is necessary that the teacher data is configured such that the past real estate image data is associated with output data (for example, a list of lines in the image data, a list of retouch items necessary for adjusting the image data) generated using a corresponding learning model.

[0033] Specifically, the generation of the learning model by the central processing unit 21 is performed by learning the image feature patterns of the past real estate image data. Such a learning model is, for example, (1) Classify past real estate image data into groups (for example, the exterior and interior of a building) with similar real estate attributes and compositions. · Calculate the image quality and composition of the past real estate image data belonging to the same group for each retouch item (for example, "color temperature" or "image tilt") to obtain representative feature amounts (hereinafter also referred to as "image features"). (2) Analyze (grasp) the tendency of the image features in the group for each retouch item based on the feature amounts of the obtained past real estate image data. It is possible to obtain such a learning procedure. Note that such analysis can be strengthened with the cooperation of experts. In particular, it is important for the learning model to collect appropriate teacher data based on the above-described analysis. Such teacher data varies depending on the purpose of use and the like. For example, learning is performed according to the purpose using the image features (feature amounts) of a plurality of past real estate image data. Note that the above-described machine learning can be performed using a program of a known neural network or the like.

[0034] For example, when learning "white balance (color temperature)" as a retouch item, it can be performed according to the following procedure. (1) First, obtain both the contextual and visual details related to the image data from past real estate image data (models). This process is performed to obtain the feature quantities of the real estate image data. Examples of such cases include "image content" such as whether it is indoors or outdoors, or whether it is a living room or a bedroom. Examples of the above-mentioned visual details include geometric elements such as lines, planes, vanishing points, and color attributes of the image such as color histogram distribution, temperature, and primary colors. (2) Next, analyze the feature quantities of multiple groups and gather the professional rules of thumb (know-how) of experts in that field to identify the patterns of past real estate image data that require specific retouching items (e.g., "white balance"). For example, in the case of "white balance", not only visual details such as "color temperature" but also contexts such as "indoors" and "artificial light" can be considered. Also, when estimating image quality and composition, it is possible to set rules such as when there is artificial lighting in an indoor image and the color temperature is high, adjustment of the white balance is required. (3) Then, based on the feature quantities obtained from the past real estate image data, generate a function that associates (maps) the original color temperature with the color temperature of the adjusted image data for the retouching item "color temperature". Such a function can be obtained, for example, by quantifying the feature quantities and finding an appropriate mathematical expression. As a result, it becomes possible to determine the recommended color temperature and recommended image values (recommended image value Iv1) for adjusting the "white balance" based on the value of "color temperature" obtained in the procedure of (1) above. Consequently, the input input image data Ii can be adjusted to an appropriate image so that it has the same "color temperature" as the past real estate image data.

[0035] Note that the generation of the learning model by the central processing unit 21 is not limited to learning the relationship between past real estate image data and the retouch item Ri. For example, it can also be used for more basic purposes by extracting the situation of the image data, detecting available furniture, clustering the images into groups to estimate the intensity of the feature amount, etc. Generally, the above-mentioned function only associates past image data with a desired result, and such analysis can improve the efficiency of the learning model generation process.

[0036] In addition to the above-mentioned learning model, a non-learning model can also be generated by the same method (approach). Examples of such models include, for example, a model based on the estimation of the vanishing point, an image data graphical model, and a model based on the estimation of image statistics. In addition, other models constructed by analyzing data other than data and extracting the knowledge of experts can also be considered.

[0037] (Memory unit 22) The memory unit 22 is, for example, a cloud-based infrastructure consisting of a cluster of so-called cloud instances. In its memory area, an image adjustment program 22A, a past real estate image database 22B, a learning model database 22C, and a retouch item database 22D are stored. The computing workload is distributed among the cloud instances, and each instance is assigned a part of the task to be processed.

[0038] The image adjustment program 22A is a program that controls the basic operation of the real estate image adjustment system 1. Specifically, the image adjustment program 22A is programmed with program instructions such as the above-mentioned learning model generation process, image estimation process ("step S202" in FIG. 5), recommended image value calculation process ("step S203" in FIG. 5), image automatic adjustment data generation process ("step S204" in FIG. 5), various data transmission processes ("step S205" in FIG. 5), etc. for the central processing unit 21 to execute.

[0039] In addition to the command procedures described above, the image adjustment program 22A also includes various programs that assist in extracting the image features of the input real estate image data and facilitate its analysis in both the image quality estimation process and the process of recommending retouching items (see FIG. 5) described later. Examples of such programs include a program for assisting in reconstructing the geometric model of an image (vanishing point estimation program). With such a program, it becomes possible to identify the vanishing point using visual details such as the lines and planes described above. As a result, when the vanishing point is far away (in the direction of infinity), the lines in the image can be determined to be parallel (not tilted), making it easier to grasp the composition of the image data. Consequently, it becomes possible to more accurately determine the presence or absence of proposals when correcting the composition.

[0040] The past real estate image database 22B is a database that stores a plurality of past real estate image data. The past real estate image database 22B stores, for example, a plurality of past real estate image data with different compositions, shooting angles, etc. for each attribute of the real estate (such as the exterior and interior views of the building). As described above, the past real estate image data is composed of various related data, each contributing to different processes when adjusting the image. For example, the image data and the list of lines are for generating a line detection learning model. In this case, it becomes possible to estimate the vanishing point and propose parameters for retouching the item "parse" as guidelines.

[0041] The learning model database 22C is a database that stores the learning models generated by performing the above-described learning model generation process. Such a learning model can be composed of, for example, a combination of a program for executing a mapping function (such as a function for associating the original color temperature with the color temperature of the adjusted image data) and learning parameters.

[0042] As shown in FIG. 3, the retouch item database 22D stores retouch item data Ri used when adjusting the image of the image automatic adjustment data Ia on the user terminal 10 side. The retouch item data Ri is stored as a program that defines an algorithm for operating on the image data to obtain desired adjusted image data. Such retouch item data Ri is stored as a program that defines an algorithm for operating on the image data to obtain desired adjusted image data, and includes, for example, "white balance (color temperature)", "white balance (color cast)", "color tone (exposure)", "color tone (contrast)", "effect (intensity)", "brightness (operation of exposure and contrast)", "perspective correction (Perspective Correction, operation of guidelines)", "privacy control (Privacy Control, operation of strength of effect)".

[0043] In addition, among the commands for adjusting the image quality and composition that are the basis of the image adjustment program, and the retouch items for providing algorithms, there are those of the type that map a predetermined color to another color (for example, white balance), as well as more complex ones, for example, those related to "perspective" that depend on a plurality of elements. Such retouch items are determined based on image features (feature quantities) extracted from the image data, such as whether the image was taken from the front or the back by applying either an affine transformation or a warp transformation, and whether there is an object to be focused on that needs to be saved, and are for correcting such problems.

[0044] (Display unit 23, input unit 24, communication unit 25) The display unit 23 is composed of, for example, a liquid crystal display (LCD: Liquid Crystal Display), and is a device that displays a predetermined image based on a command from the central processing unit 21. The input unit 24 consists of, for example, a known keyboard, mouse, and touch panel, and is a device used when adding, changing, deleting, etc. various data stored in the storage unit 22. The communication unit 25 is an interface capable of communicating with communication devices such as the user terminal 10 via the public communication line 2.

[0045] <Control Processing in the Real Estate Image Adjustment System 1> Next, the control processing in this real estate image adjustment system 1 will be described with reference to FIGS. 1 to 8. In the following, the control processing in the user terminal 10 and the control processing in the server 20 will be separately described. Also, in the following, for the sake of convenience of explanation, it is assumed that the user terminal 10 and the server 20 are in an interlocking state, the input image data Ii is captured by the user terminal 10, and the learning model is stored in the learning model database 22C of the server 20.

[0046] (Control Processing in the User Terminal 10) First, the control processing in the user terminal 10 will be described with reference to FIGS. 1, 2, 4, and 6 to 8.

[0047] (Step S101) As shown in FIG. 4, the control processing in the user terminal 10 mainly starts with the central processing unit 11 performing the processing of step S101. The central processing unit 11 performs a process of determining whether the input image data Ii (see FIG. 6) has been transmitted to the server 20 in step S101. When the central processing unit 11 determines that the input image data Ii has been transmitted, it transfers the process to step S102, and when it determines that it has not been transmitted, it repeatedly executes this step S101. In the following, the case where the input image data Ii is tilted to the left as shown in FIG. 6 will be taken as an example for explanation.

[0048] (Step S102) As shown in FIG. 4, the central processing unit 11 performs a process of determining whether or not it has received various data transmitted from the server 20. This various data is data including image automatic adjustment data Ia (see FIG. 7), retouch item data Ri (see FIG. 8), and recommended image value Iv1 (see FIG. 8) described later, and is transmitted to the user terminal 10 by the process of step S205 (control process in the server 20 in FIG. 5). When the central processing unit 11 determines that it has received the image automatic adjustment data Ia, the process proceeds to step S103, and when it determines that it has not received it, the process of this step S102 is repeatedly executed.

[0049] (Step S103) In step S103, the central processing unit 11 performs a process (image comparison display process) of displaying the image automatic adjustment data Ia (see FIG. 8) on the display unit 13. This image automatic adjustment data Ia is data in which the composition such as the image quality and the inclination of the input image data Ii is adjusted, and is generated by the process of step S204 (control process in the server 20 in FIG. 5).

[0050] Specifically, the central processing unit 11 executes control to display an image (hereinafter referred to as "recommended image") as shown in FIG. 7 on the display unit 13. In the example shown in FIG. 7, "input image" is displayed at the upper left of the screen, "image estimation result" is displayed at the lower left of the screen, "automatically adjusted image" is displayed at the upper right of the screen, and "image automatic adjustment content" is displayed at the lower right of the screen on the display unit 13, respectively. The "input image" is the input image data Ii transmitted to the server 20 in step S101. The "image estimation result" is an estimation result (analysis result) estimated by the process of step S202 (control process in the server 20 in FIG. 5). Such "image estimation result" can be, for example, "image inclination" adjustment required, "white balance" adjustment required, "color tone" adjustment required, "effect" adjustment required (see FIG. 7).

[0051] The "automatically adjusted image" is the automatically adjusted image data Ia in which the composition such as the image quality and the inclination of the input image data Ii is automatically adjusted by the central processing unit 21 of the server 20, and is displayed on the display unit 13 side by side with the input image data Ii on the right. As a result, the user can easily compare the input image data Ii with the automatically adjusted image data Ia, and thus can easily grasp the quality of the automatically adjusted image data Ia. The "details of image automatic adjustment" shows the specific details when the input image data Ii is adjusted to the automatically adjusted image data Ia. Such "details of image automatic adjustment" can be, for example, contents such as "image inclination adjusted", "white balance adjusted", "color tone adjusted", "effect adjusted" (see Fig. 7).

[0052] As shown in Fig. 4, after performing the image comparison display process, the central processing unit 11 transfers the process to step S104.

[0053] (Step S104) In step S104, the central processing unit 11 performs a process of determining whether there is a selection operation by the user. For example, in the case of the example shown in Fig. 7, the central processing unit 11 performs a process of determining whether either the icon of "OK with this" or the icon of "image editing display" has been selected by the user. In the example shown in Fig. 7, when further adjusting the image of the automatically adjusted image data Ia, the icon of "image editing display" is selected, and when no further adjustment is required, the icon of "OK with this" is selected. As shown in Fig. 4, when the central processing unit 11 determines that there is a selection operation by the user, it transfers the process to step S105, and when it determines that there is no selection operation, it repeatedly executes the process of this step S104.

[0054] (Step S105) In step S105, the central processing unit 11 performs a process of determining whether the image adjustment work has been completed. Specifically, when the "OK" icon is selected in step S104, the central processing unit 11 determines that the image adjustment operation is completed, and when the "Image Editing Display" icon is selected, it performs a process of determining that the image adjustment operation is not completed. When the central processing unit 11 determines that the image adjustment operation is completed, it transfers the process to step S110, and when it determines that the image adjustment operation is not completed, it transfers the process to step S106.

[0055] (Step S106) In step S106, the central processing unit 11 performs a process (image editing display process) of displaying the retouch item data Ri, the recommended image value Iv1, and the user image setting value Iv2 on the display unit 13.

[0056] As shown in FIG. 8, the retouch item data Ri is composed of a recommended image value Iv1 (for example, "white balance (color temperature)") recommended by the process of step S203 (control process in the server 20 in FIG. 5). The recommended image value Iv1 is a parameter value recommended by the process of step S203 (control process in the server 20 in FIG. 5), and indicates retouch items (for example, "white balance (color temperature)", "white balance (color cast)") necessary to adjust the input image data Ii to the recommended image. The user image setting value Iv2 is a value (value) displayed on the display unit 13 when the parameter value of each retouch item data Ri is adjusted (when desired) from the recommended image value Iv1.

[0057] Specifically, the central processing unit 11 executes control to switch the display on the display unit 13 from the image shown in FIG. 7 (hereinafter referred to as the "first image") to the image shown in FIG. 8 (hereinafter referred to as the "second image"). In the example shown in FIG. 8, the real estate image data being adjusted is displayed on the left side of the screen, the display column for the retouch item data Ri is on the right side of the screen, and the icon “OK with this” is displayed on the lower side of the screen in the display unit 13, respectively. In this embodiment, when the first image is switched and displayed to the second image, image data adjusted to the recommended image value Iv1 (image automatic adjustment data Ia (image data obtained by adjusting the input image data Ii to the recommended image)) is configured to be displayed as the default image (see FIG. 8). Thereby, the user can perform image adjustment based on the automatically adjusted image data. Note that FIG. 8 shows the state immediately after the user changes (adjusts) the parameter value of each retouch item data Ri from the recommended image value Iv1 to a desired value (user image setting value Iv2). Thus, it can be said that the recommended image value Iv1 functions as a reference value in which the parameter values of each retouch item data Ri are automatically adjusted by the server 20.

[0058] In the display column for the retouch item data Ri, the retouch items necessary for adjusting the image quality and composition of the input image data Ii to the recommended image are listed. For example, in the second image of FIG. 8, as retouch items for adjusting the recommended image to the image and composition desired by the user, “White Balance (Color Temperature)”, “White (Color Cast)”, “Color Tone (Exposure)”, “Color Tone (Contrast)”, and “Effect (Intensity)” are displayed.

[0059] As described above, the retouch item data Ri includes a retouch item name and a horizontal slider having a knob icon Pi. The parameter value of the retouch item data Ri can be increased or decreased by moving the knob icon Pi to the left or right.

[0060] Around the horizontal slider, the current parameter value of the retouch item data Ri is displayed (the “parameter display unit Pv”), and the recommended image value Iv1 and the user image setting value Iv2 are appended. For example, in the second image of FIG. 8, in the case of "White Balance (Color Temperature)" (retouch item data Ri), in the initial state (the state immediately after switching from the first image to the second image), the recommended image value Iv1 is "4000". If the parameter value of "White Balance (Color Temperature)" is to be changed, for example, the knob icon Pi may be moved leftward until the user image setting value Iv2 (the value of the parameter display part Pv) becomes "2635".

[0061] In this embodiment, when the knob icon Pi is moved to increase or decrease the parameter value of the retouch item data Ri (for example, "White Balance (Color Temperature)"), accordingly, the image of the real estate image data on the left side of the screen (for example, "White Balance (Color Temperature)") is also configured to change continuously.

[0062] As shown in FIG. 4, after performing the second image display process, the central processing unit 11 moves the process to step S107. In this embodiment, as the display modes of the data indicating the recommended image value Iv1 and the user image setting value Iv2, two examples are given: a specific number (for example, "4000") and a vertical line (see FIG. 8) representing its position. However, only one of them may be used, or other modes (for example, a figure) may also be acceptable.

[0063] (Step S107) In step S107, the central processing unit 11 performs a process of determining whether there is a predetermined operation by the user. Specifically, in the case of the example shown in FIG. 8, the central processing unit 11 performs a process of determining whether there is either an operation of moving the knob icon Pi or an operation of selecting the "OK with this" icon. In the example shown in FIG. 8, when there is an operation of selecting the "OK with this" icon, the work of adjusting the input image data Ii is configured to end. As shown in FIG. 4, when the central processing unit 11 determines that there is a predetermined operation, it moves the process to step S108, and when it determines that there is no predetermined operation, it repeatedly executes the process of this step S107.

[0064] (Step S108) In step S108, the central processing unit 11 performs a process of determining whether the image adjustment operation has been completed. For example, in the case of the example shown in FIG. 8, when an operation of selecting the "OK" icon is performed in step S107, the central processing unit 11 determines that the image adjustment operation has been completed. When an operation of moving the knob icon Pi is performed, the central processing unit 11 performs a process of determining that the image adjustment operation has not been completed. When the central processing unit 11 determines that the image adjustment operation has been completed, the process proceeds to step S110. When the central processing unit 11 determines that the image adjustment operation has not been completed, the process proceeds to step S109.

[0065] (Step S109) In step S109, the central processing unit 11 performs a process (image data change display process) of changing the image quality and composition of the real estate image data (image automatic adjustment data Ia) and displaying it on the display unit 13. This step S109 is executed when, in step S108, an operation of moving the knob icon Pi in the left - right direction is performed, that is, when it is determined that an operation of increasing or decreasing the parameter value of a predetermined retouch item data Ri (for example, "white balance (color temperature)") has been performed. Therefore, in step S109, when the parameter value of the retouch item data Ri is increased or decreased, the central processing unit 11 continuously changes the image (image quality) of the real estate image data (for example, "white balance (color temperature)") displayed on the left side of the screen in FIG. 8 according to the amount of increase or decrease. As a result, the user can operate the knob icon Pi (increase or decrease the parameter value of the retouch item data Ri) while confirming the process of the changing real estate image data, and thus can adjust the image to suit their preferences. After performing the image data change display process, the central processing unit 11 returns the process to step S107.

[0066] (Step S110) In step S110, the central processing unit 11 performs a process of transmitting the adjusted real estate image data to the server 20 (adjusted image data transmission process). This step S110 is executed when it is determined in the processes of steps S105 and S108 that the image adjustment operation has been completed, that is, when the image of the real estate image data is not adjusted any further. As a result, on the server 20 side, by comparing the real estate image data before and after image adjustment, it is possible to machine-learn what kind of image adjustment the user has performed, and it becomes possible to update the learning model stored in the learning model database 22C (see FIG. 3) at any time ( "step S207" in FIG. 5). As a result, the accuracy of various processes performed using the learning model (for example, the "image estimation process" performed in "step S202" in FIG. 5) can be improved, so that it becomes possible to provide the user with more appropriate real estate image data of the image. Note that in this embodiment, in this process, in addition to the adjusted image data, data regarding the parameter values (user image setting value Iv2) of each retouch item adjusted by the user in the process of step S107 is also configured to be transmitted to the server 20. After performing the adjusted image data transmission process, the central processing unit 11 ends the control process in the present user terminal 10.

[0067] (Control Process in Server 20) Next, the control process in the server 20 will be described with reference to FIGS. 1, 3, and 5.

[0068] (Step S201) As shown in FIG. 5, the control process in the server 20 mainly starts with the central processing unit 21 performing the process of step S201. In step S201, the central processing unit 21 performs a process of determining whether or not the input image data Ii (see FIG. 6) has been received. This input image data Ii is transmitted by the process of step S101 (control process in the user terminal 10 in FIG. 4). When the central processing unit 21 determines that the input image data Ii is being received, it transfers the process to step S202. When it determines that the input image data Ii is not being received, it repeatedly executes this step S201.

[0069] (Step S202) In step S202, the central processing unit 21 performs a process of estimating the image of the input image data Ii (image estimation process). Specifically, the central processing unit 21 uses the learning model stored in the learning model database 22C to perform a process of estimating the image of the input image data Ii. Note that the above image estimation process corresponds to the "image estimation step" and "image estimation process" described in the claims.

[0070] The image estimation process can be performed, for example, according to the following procedure. (1) Using the learning model (recommended image value Iv1), calculate the image features (feature amounts) of the input image data Ii. Such calculation of image features can be efficiently performed because the feature amounts are specified for each retouch item at the time of generating the learning model. Examples of such image features include visual details such as "indoors" and "artificial light" when the retouch item is "white balance". (2) Search for the group to which the input image data Ii belongs, and extract the representative image (recommended image value Iv1 (image features)) of that group. (3) Based on the image extracted in (2) above, estimate for each image retouch item of the input image data Ii.

[0071] In this embodiment, past real estate image data is stored in the storage unit 22 together with its retouch items (see FIG. 3 and the like). As a result, when the group to which the input image data Ii belongs is searched, the retouch item data Ri necessary for image adjustment of the image data within the group can be extracted, and the recommended data (recommended retouch item data Ri) can be extracted.

[0072] For specific retouch item data Ri, a non-learning model program involving logical inference can be used without using a learning model. In this case, it is also possible to store the program in the image adjustment program 22A. Examples of such retouch items include items related to defects in the composition of an image, such as "Perspective". Generally, for such items, in order to adjust basic parts such as lines and planes that make up the skeleton of the image, when making a judgment such as whether such adjustment is necessary based on a machine-learned learning model, it is often the case that an appropriate image (for example, a graphical model of the image starting from the vanishing point) cannot be obtained. In such a case, for example, when specifying the vanishing point, it is also possible to use a non-learning estimator that executes an algorithm on the detected lines and planes. Thereby, it is possible to adjust defects in the composition of the image based on the position of the vanishing point. In addition, when there are defects such as when the vanishing point is not far enough away or when an orthogonal structure is not formed, so-called perspective correction can also be performed.

[0073] In the present embodiment, by performing the above-described image estimation process, it becomes possible to list up the retouch items necessary to adjust the input image data Ii to the recommended image and the recommended image value Iv1 thereof. Note that such a result (image estimation result) can be displayed in the same manner as the "image estimation result" of the first image shown in FIG. 7.

[0074] After performing the image estimation process, the central processing unit 21 transfers the process to step S203.

[0075] (Step S203) In step S203, the central processing unit 21 performs a process (recommended image value calculation process) of calculating the recommended image value Iv1 (see FIG. 8). As described above, the recommended image value Iv1 is a parameter value of the retouch item necessary to adjust the image of the input image data Ii to the recommended image. As described above, the recommended image can be formed using the learning model stored in the learning model database 22C.

[0076] For example, when the data (data necessary to adjust the input image data Ii to the recommended image) listed or the like in the process of step S202 is "White balance (color temperature): 4000", "White balance (color cast): 15", "Color tone (exposure amount): 48", "Color tone (contrast): 23", "Effect (intensity): 8", the central processing unit 21 performs a process of setting these to the recommended image values Iv1 of each retouch item (see FIG. 8). Thereby, it becomes possible to obtain a recommended image (composite image) as shown in FIGS. 7 and 8. Note that the process of selecting and recommending each retouch item such as the above "white balance" and "color tone" corresponds to the "retouch item data recommendation step" and the "retouch item data recommendation step" described in the claims.

[0077] In this case, all of the retouch items listed or the like may be changed, or only specific retouch items may be changed. For example, in the case of a retouch item related to perspective, since it is necessary to determine the degree of defect in the composition of the input image data Ii, for example, if the input image data Ii is an image of a real estate taken from the front, it is composed of two orthogonal vanishing points at infinity, so the degree of defect is low. In this regard, it can be said that it is preferable to configure such a retouch item to correct both axes corresponding to the two vanishing points. Note that if configured in this way, if the input image data Ii is an image of a real estate taken from the back side, one axis may be greatly inclined, and there are many cases where it is difficult to correct this. Such a problem can be configured to correct only the one axis.

[0078] After performing the recommended image value calculation process, the central processing unit 21 moves the process to step S204.

[0079] (Step S204) In step S204, the central processing unit 21 performs image automatic adjustment data generation processing. Specifically, the central processing unit 21 performs processing to automatically adjust the image of the input image data Ii based on the recommended image value Iv1 calculated in step S203. For example, when the recommended image value Iv1 is "White balance (color temperature): 4000", "White balance (color cast): 15", "Color tone (exposure amount): 48", "Color tone (contrast): 23", "Effect (intensity): 8" (see FIG. 8), the central processing unit 21 performs processing to adjust the parameter values of each retouch item of the input image data Ii to these values. Thereby, it becomes possible to automatically adjust the input image data Ii to the recommended image.

[0080] As described above, when there is a problem with elements other than image quality (for example, composition such as the inclination or size of the image) in the central processing unit 21 according to the present embodiment, it is configured to also perform processing to detect and automatically adjust this. Therefore, when the input image data Ii is inclined as shown in FIG. 6, the central processing unit 21 also performs processing to automatically correct the inclination (or perspective) in this step S204 (see FIG. 7).

[0081] After performing the image automatic adjustment data generation processing, the central processing unit 21 moves the processing to step S205.

[0082] (Step S205) As shown in FIG. 5, in step S205, the central processing unit 21 performs processing to transmit various data to the user terminal 10 (various data transmission processing). These various data are composed of at least the recommended image value Iv1 calculated in the recommended image value calculation processing (step S203), the retouch item data Ri corresponding thereto, and the image automatic adjustment data Ia generated in the image automatic adjustment data generation processing (step S204). As a result, on the display unit 13 of the user terminal 10, in the first image (see FIG. 7), the image automatic adjustment data Ia is displayed (image comparison process (“step S103” in FIG. 4)), and in the second image (see FIG. 8), the retouch item data Ri and the recommended image value Iv1 are displayed (image editing display process (“step S106” in FIG. 4)). After performing various data transmission processes, the central processing unit 21 transfers the process to step S206. Note that the above various data transmission processes correspond to the “display step” and “display process” described in the claims.

[0083] (Step S206) In step S206, the central processing unit 21 performs a determination process to determine whether various data has been received. This data is the data transmitted in the process of step S110 (control process by the user terminal 10 in FIG. 4) when the user has completed the image adjustment of the real estate image data, and is mainly composed of the adjusted image data and the user image setting value Iv2 (see FIG. 8) when the user changes the parameter value. When it is determined that the central processing unit 21 has received various data such as the adjusted image data, the process is transferred to step S207. When it is determined that the data has not been received, this step S206 is repeatedly executed.

[0084] (Step S207) In step S207, the central processing unit 21 performs a learning model update process. For example, the central processing unit 21 (1) Stores the adjusted image data adjusted by the user (see “step S108” in FIG. 4, etc.) in the past real estate image database 22B (see FIG. 3). (2) Generates (updates) a learning model taking this data into account at the timing when a predetermined amount of data is stored in the past real estate image database 22B. (3) Stores the learning model generated in (2) above in the learning model database 22C (see FIG. 3). The processing is performed in the following procedure. In the processing of (1) above, in order to grasp the degree of image adjustment, the input image data Ii (see "Step S101" in FIG. 4 and "Step S201" in FIG. 5, etc.) corresponding to the adjusted image data, and the user image setting value Iv2 are preferably configured to be stored in the past real estate image database 22B.

[0085] As a result, a further learned learning model can be constructed, so that it is possible to improve the estimation accuracy of the input image data Ii obtained in the image estimation process (Step S202) and the calculation accuracy of the recommended image value Iv1 in the recommended image value calculation process (Step S203). As a result, it is possible to provide the user with more optimal real estate image data of the image.

[0086] After performing the learning model update process, the central processing unit 21 ends the control process in this server 20.

[0087] As described above, in this embodiment, by simply transmitting the input image data Ii to the server 20, it is possible to obtain the image automatic adjustment data Ia automatically adjusted to an appropriate image, and also to obtain the retouch item data Ri and the recommended image value Iv1 for adjusting to a more appropriate image. Therefore, according to this embodiment, even a user who is not used to image adjustment can easily and quickly adjust the real estate image data to a more appropriate image. In addition, as described above, since the retouch item data Ri and the recommended image value Iv1 are based on the past real estate image data captured by a professional cameraman, etc., it is possible to present the user with an index for image adjustment with a certain standard.

[0088] In this embodiment, the server 20 is configured to detect and automatically adjust defects in elements other than image quality (e.g., the tilt and size of an image). However, such adjustment can also be configured to be performed on the user terminal 10 side. In this case, only the detection of defects in elements other than image quality is performed on the server 20 side, and similar to the retouch item data Ri, it is possible to display data (item name and horizontal slider) indicating the defect on the user terminal 10 (the "second image" in FIG. 8).

[0089] Also, in this embodiment, the server 20 is configured to generate the image automatic adjustment data Ia and display it on the display unit 13 of the user terminal 10. However, such a configuration can also be omitted.

[0090] Furthermore, in this embodiment, on the server 20 side, processes such as learning model generation processing, image estimation processing (step S202), recommended image value calculation processing (step S203), image automatic adjustment data generation processing (step S204), various data transmission processing (step S205), learning model update processing (step S207), etc. are performed. However, each of these processes can also be performed on the user terminal 10. In this case, the storage unit 12 of the user terminal 10 stores the information stored in the storage unit 22 of the server 20, and the control processing in the server 20 shown in FIG. 5 may be performed by the central processing unit 11 of the user terminal 10. With such a configuration, it is possible to omit the server 20, so that the configuration can be simplified. Note that when configured in this way, various data such as the retouch item data Ri and the recommended image value Iv1 may be appropriately provided from the server 20 to the user terminal 10, and it is also possible to configure the user terminal 10 and the server 20 to cooperate in order to grasp the user's preferences. The user terminal 10 that executes the control processing in the server 20 corresponds to the "real estate image adjustment device" described in the claims.

[0091] In addition, in this embodiment, the retouch item data Ri and the recommended image quality value Iv1 are displayed on the display unit 13 of the user terminal 10, but it is also possible to display only the retouch item data Ri (see FIG. 8). Note that in the example shown in FIG. 8, the state where the user has changed each retouch item data Ri from the recommended image value Iv1 is shown, and at that time, the changed value (see "user image setting value Iv2" in FIG. 8) is configured to be displayed, but it is also possible to configure not to display this.

[0092] As described above, the embodiments to which the invention made by the present inventor is applied have been described. However, the present invention is not limited by the description and drawings that form a part of the disclosure of the present invention according to this embodiment. That is, it should be added that all other embodiments, examples, operation techniques, etc. made by those skilled in the art based on this embodiment are of course included in the scope of the present invention.

Explanation of Reference Numerals

[0093] 1 Real estate image adjustment system 2 Public communication line 10 User terminal 10A Personal computer 10B Mobile terminal 11 Central processing unit 12 Storage unit 12A Image editing program 13 Display unit 14 Input unit 15 Communication unit 20 Server 21 Central processing unit 22 Storage unit 22A Image adjustment program 22B Past real estate image database 22C Learning model database 22D Retouch item database 23 Display unit 24 Input unit 25 Communication unit Ii Input image data Ia Image automatic adjustment data Ri Retouch item data Pi Knob icon Pv Parameter display section Iv1 Recommended image value Iv2 User image setting value

Claims

1. A real estate image adjustment system comprising a user terminal having input means capable of inputting real estate image data obtained by imaging a real estate, and a server for adjusting the image quality and composition of the real estate image data transmitted from the user terminal, wherein the server has storage means for storing a plurality of retouch item data indicating adjustment items for the image quality and composition of the real estate image data; image estimation means for estimating the image quality and composition of the real estate image data transmitted from the user terminal using a learning model machine-learned with the past real estate image data as teacher data; retouch item data recommendation means for selecting and recommending, from the storage means, the retouch item data necessary for adjusting the image quality and composition of the real estate image data estimated by the image estimation means to a recommended image based on the learning model; image adjustment means for adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by the retouch item data recommendation means; transmission means for transmitting the adjusted real estate image data adjusted by the image adjustment means to the user terminal together with the recommended retouch item data; and the user terminal is provided with display means for displaying the adjusted real estate image data and the recommended retouch item data transmitted by the transmission means. A real estate image adjustment system.

2. The input means is capable of adjusting the recommended retouch item data transmitted from the transmission means, and the image adjustment means adjusts the image quality and composition of the real estate image data based on the adjusted real estate image data and the adjusted retouch item data input by the input means. The real estate image adjustment system according to claim 1.

3. The server further includes recommended image value calculation means for calculating a recommended image value corresponding to the recommended image for each of the retouch item data, the transmission means transmits data indicating the recommended image value for each of the retouch item data to the user terminal, and the display means displays the data indicating the recommended image value transmitted by the transmission means together with the retouch item data. The real estate image adjustment system according to claim 1 or 2.

4. A real estate image adjustment device capable of adjusting the image quality and composition of real estate image data obtained by imaging a real estate, comprising input means capable of inputting the real estate image data, Storage means for storing a plurality of retouch item data indicating adjustment items for the image quality and composition of the real estate image data; Image estimation means for estimating the image quality and composition of the real estate image data input by the input means using a learning model machine-learned using the past real estate image data as teacher data; Retouch item data recommendation means for selecting and recommending, from the storage means, the retouch item data necessary for adjusting the image quality and composition of the real estate image data estimated by the image estimation means to a recommended image based on the learning model; Image adjustment means for adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by the retouch item data recommendation means; Display means for displaying the adjusted real estate image data adjusted by the image adjustment means together with the recommended retouch item data; A real estate image adjustment device comprising:

5. A real estate image adjustment method capable of adjusting the image quality and composition of real estate image data obtained by imaging a real estate, comprising: An input step of inputting the real estate image data; An image estimation step of estimating the image quality and composition of the real estate image data input by performing the input step using a learning model machine-learned using the past real estate image data as teacher data; A retouch item data recommendation step of selecting and recommending, from storage means, the retouch item data necessary for adjusting the image quality of the real estate image data estimated by performing the image estimation step to a recommended image based on the learning model; An image adjustment step of adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by performing the retouch item data recommendation step; A display step of causing display means to display the adjusted real estate image data adjusted by performing the image adjustment step together with the recommended retouch item data; A real estate image adjustment method including:

6. A real estate image adjustment program capable of adjusting the image quality and composition of real estate image data obtained by imaging a real estate, comprising: An image estimation step of estimating the image quality and composition of the real estate image data input from input means using a learning model machine-learned using the past real estate image data as teacher data; A retouch item data recommendation step of selecting and recommending, from storage means, retouch item data necessary for adjusting the image quality and composition of the real estate image data estimated by performing the image estimation step to a recommended image based on the learning model; An image adjustment step of adjusting the image quality and composition of the real estate image data based on the recommended retouch item data recommended by performing the retouch item data recommendation step; A display step of causing a display means to display the adjusted real estate image data adjusted by performing the image adjustment step together with the recommended retouch item data; A real estate image adjustment program for causing a computer to execute the above steps.

Images