Imaging device, control method, and program

Abstract

This invention provides an imaging device, control method, and program that allow even inexperienced users to easily obtain photographs that match their intentions. [Solution] The imaging device 1000 converts voice data of the user's request for shooting effect, acquired via a microphone, into a prompt. Using a machine learning algorithm, it generates output data consisting of shooting parameters that maximize the shooting effect and a suggested shooting image, based on the input data including the live view image, its shooting parameters, and the prompt. It then displays, in a switchable manner, at least the live view image with zero shooting effect and the suggested shooting image with maximum shooting effect. In response to a first user operation, it adjusts the shooting effect within a range from zero to maximum. Upon a second user operation, it performs shooting preparation using the adjusted shooting parameters.

Description

Technical Field

[0001] The present invention relates to an imaging device, a control method, and a program, and more particularly to an imaging device, a control method, and a program having a shooting assist function that utilizes a machine learning algorithm to assist shooting settings by a user.

Background Art

[0002] In recent years, in imaging devices such as digital cameras, in order for a user to perform optimal shooting settings according to the shooting scene, the user has had to manually adjust many parameters such as shutter speed, aperture, and ISO sensitivity. Therefore, it has been difficult for users who are not used to shooting to perform shooting as intended. There is an automatic mode (for example, "program auto function") in which the imaging device determines the above many parameters, but this mode does not guarantee optimal results in all situations.

[0003] On the other hand, Patent Document 1 discloses a learned model that outputs the shooting scene of a captured image input by an imaging device, recommended camera setting items for the captured scene and its determined shooting scene, and the adjustment range thereof. Thereby, the user can easily perform shooting settings according to the shooting scene by adjusting the shooting parameters within the adjustment range.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the technology described in Patent Document 1, only the recommended camera settings can be adjusted by the user. Therefore, if the shooting scene determination result by the trained model does not match the user's preferences, it may be difficult for the user to achieve the desired shooting settings. Furthermore, if there are multiple camera settings that need adjustment, the user will be forced to perform complex shooting setting operations.

[0006] Therefore, the present invention aims to provide an imaging device, a control method, and a program that allow even users unfamiliar with photography to easily obtain photographs that match their intentions. [Means for solving the problem]

[0007] To solve the above problems, the imaging device according to claim 1 of the present invention is characterized by comprising: an imaging means; a request acquisition means for acquiring a request for an imaging effect from a user; a conversion means for converting the request for an imaging effect into a prompt; a first generation means for generating, as output data, imaging parameters that maximize the imaging effect and a suggested imaging image, using a machine learning algorithm from input data including a live view image obtained by the imaging means, imaging parameters associated with the live view image, and the prompt; an image display means for switching between displaying at least the live view image with zero imaging effect and the suggested imaging image with maximum imaging effect; an adjustment means for adjusting the imaging effect within a range from zero to maximum in response to a first user operation; a second generation means for generating the imaging parameters that result in the adjusted imaging effect using the machine learning algorithm; and a imaging preparation means for performing imaging preparation using the imaging parameters of the adjusted imaging effect when a second user operation is received.

[0008] To solve the above problems, the imaging device according to claim 15 of the present invention is an imaging device connected to a generation device, the imaging device comprising: an imaging means; a request acquisition means for acquiring a request for an imaging effect from a user; a conversion means for converting the request for an imaging effect into a prompt; a first data transmission / reception unit that transmits a first communication data including a live view image obtained by the imaging means, an imaging parameter associated with the live view image, and the prompt to the generation device, and receives from the generation device an imaging parameter and imaging suggestion image that maximize the imaging effect, generated from the first communication data using a machine learning algorithm; and at least the imaging device The system is characterized by comprising: an image display means that can switch between displaying a live view image with zero shadow effect and a shooting plan image with maximum shooting effect; an adjustment means that adjusts the shooting effect within a range from zero to maximum in response to a first user operation; a second data transmission / reception unit that transmits a shooting effect value indicating the magnitude of the adjusted shooting effect to the generation device and receives from the generation device the shooting parameters and shooting plan image of the adjusted shooting effect generated from the shooting effect value using the machine learning algorithm; and a shooting preparation means that, when a second user operation is performed, performs shooting preparation using the shooting parameters of the adjusted shooting effect. [Effects of the Invention]

[0009] According to this invention, even users unfamiliar with photography can easily take photos exactly as they intend. [Brief explanation of the drawing]

[0010] [Figure 1A] This is a schematic cross-sectional view of an imaging device according to the first embodiment of the present invention. [Figure 1B] Figure 1A is a block diagram showing the hardware configuration of the imaging device. [Figure 2] Figure 1A is an external view of the imaging device. [Figure 3] This is a flowchart of the setting change process according to the first embodiment of the present invention. [Figure 4]This is a diagram illustrating the image adjustment mode for a proposed image according to the first embodiment of the present invention. [Figure 5] Figure 1B shows the protocol generation unit and an example of the generation unit's configuration. [Figure 6] This diagram shows a modified configuration of the generation unit. [Figure 7] This is a flowchart of the setting change process on the imaging device side of the imaging system according to a second embodiment of the present invention. [Figure 8] This is a flowchart of the setting change process on the generation device side of the imaging system according to the second embodiment of the present invention. [Figure 9] This is a block diagram showing the hardware configuration of an imaging device according to a second embodiment of the present invention. [Figure 10] This figure shows the configuration of an imaging system according to a second embodiment of the present invention. [Modes for carrying out the invention]

[0011] Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Note that the following embodiments do not limit the invention claimed. Furthermore, while multiple features are described in the embodiments, not all of them are essential to the invention, and the multiple features may be combined arbitrarily. In addition, in the accompanying drawings, the same or similar configurations are given the same reference numerals, and redundant descriptions are omitted.

[0012] Embodiments of the present invention will be described below with reference to the drawings.

[0013] (First example) Figure 1A is a diagram showing the configuration of an imaging device 1000 according to a first embodiment of the present invention.

[0014] As shown in Figure 1A, the imaging device 1000 consists of an imaging optical system 100 and a camera body 200.

[0015] The imaging optical system 100 includes an aperture 11, a shake correction lens group 12, and a focus lens group 13, and can guide an optical image to the camera body 200.

[0016] The camera body 200 includes an imaging element 21 (imaging means) that photoelectrically converts the optical image of the imaging optical system 100 and a mechanical shutter 22 that adjusts the exposure time. The camera body 200 also includes, as display units, a rear liquid crystal 23 (rear panel) on the back surface and a small liquid crystal 24 provided together with an optical system 25 in the upper finder unit 206 (electronic viewfinder: EVF). The images captured by the imaging element 21 can be displayed on these display units. Note that the imaging element 21 may be an imaging element having an electronic shutter function. In this case, the mechanical shutter is unnecessary, and even when the mechanical shutter is provided, when the exposure time is adjusted by the electronic shutter, the mechanical shutter remains fully open.

[0017] At the time of shooting, when the user performs a so-called "half-press", that is, lightly presses a shutter button (not shown) to the first stage, the shooting parameters such as the shutter speed and aperture value are set by the autofocus and automatic exposure mechanisms. Further, when the user performs a so-called "full-press", that is, deeply presses the shutter button from the half-press to the second stage, the mechanical shutter 22 or the electronic shutter function of the imaging element 21 operates to perform imaging.

[0018] FIG. 1B is a block diagram showing the hardware configuration of the imaging device 1000.

[0019] In addition to the above-described configuration, the camera body 200 includes an electric circuit 20 and a microphone 27.

[0020] The electric circuit 20 has a CPU 201, an image processing unit 202, a control unit 203, a generation unit 204, a prompt generation unit 205, etc. mounted thereon.

[0021] The aperture 11, the shake correction lens group 12, the focus lens group 13, and the mechanical shutter 22 are each controlled by the control unit 203 via driving means (not shown).

[0022] The signal converted photoelectrically by the image sensor 21 can be converted into digital data via the image processing unit 202 and stored on a recording medium (not shown).

[0023] The viewfinder unit 206 is further equipped with an eyepiece sensor 26, which can detect whether or not the user is looking through the viewfinder unit 206.

[0024] The prompt generation unit 205 (conversion means) generates prompts from the audio data input by the microphone 27. The generation unit 204 is equipped with a trained model such as a machine learning-based neural network, and inputs the captured image data into the trained model to generate a new image, a proposed image to be captured, and shooting parameters for capturing that proposed image. Further details will be described later.

[0025] The CPU201 is a processing unit that can electrically control all of the above elements. In Figure 1B, the control signal lines are omitted, and only the flow of information between each element is indicated by arrows.

[0026] Figure 2 is an external view of the imaging device 1000 as seen from the rear.

[0027] In Figure 2, the same reference numerals are used for the components as in Figures 1A and 1B, and redundant explanations are omitted.

[0028] The user changes the shooting parameters and takes a picture using the operation unit 29, which consists of buttons and other components attached to the imaging device 1000, and the electronic dial 28 (setting change means). If the rear LCD 23 has a touch panel function, it may be possible to change the shooting parameters by touching the rear LCD 23 (setting change means).

[0029] The user can understand the current shooting parameter settings through the display output on the rear LCD 23 and the small LCD 24. Furthermore, adjustments and changes to the shooting effect on the shooting sample image displayed on the rear LCD 23 or the small LCD 24 can be made using the electronic dial 28 or the control unit 29.

[0030] <Regarding the shooting parameters generated by the generation unit 204> The shooting parameters generated by the generation unit 204 are not particularly limited. For example, typical parameters for shooting bright images include ISO sensitivity, shutter speed, exposure, white balance, image contrast, and aperture value for adjusting the depth of field. Of course, parameters for other camera settings that affect the shooting effect of the captured image can also be included. For example, saturation, sharpness, lighting correction, tone priority, shooting style, noise reduction, color correction, and image processing settings may also be included.

[0031] Furthermore, since users have different preferences for photographs, and some may want to focus on multiple subjects or on a specific subject, information about the focus position (focus setting / subject tracking setting) may also be included in the shooting parameters.

[0032] The generation unit 204 is a processing unit having a trained model that can present suggested images with shooting effects that take into account the user's photo preferences and shooting conditions, and the trained model is composed of, for example, a multi-layer neural network.

[0033] In this embodiment, the trained model of the generation unit 204 has learned the user's preferences in advance for each lens type of the imaging optical system of multiple imaging devices, including the imaging device 1000, and for each camera type of the camera body, before shooting. Using this trained model, the unit performs inference processing with the live view image before shooting as input and generates shooting parameters and a suggested shooting image. Subsequently, the generation unit 204 outputs an adjustment screen to the rear LCD 23 or the viewfinder unit 206, which allows the user to adjust the suggested shooting image.

[0034] Furthermore, the shooting effect value x (a variable in the range of 0 to 1) which indicates the magnitude of the shooting effect of the proposed image compared to the live view image before shooting, is also input data for this trained model. When the shooting effect value x is changed from its initial value (=1) which maximizes the shooting effect, the inference process is performed again using the changed shooting effect value x. Here, the shooting effect value x is changed according to user operations on the adjustment screen, as will be described later using Figure 4.

[0035] The training method for the pre-trained model provided by the generation unit 204 will be described later.

[0036] <Processing flow from shooting plan generation by generation unit 204 to setting change> Figure 3 is a flowchart of the process for changing the settings of the imaging device 1000 in this embodiment. The following process is achieved by the CPU 201 sending commands to the control unit 203 to control each part of the imaging device 1000.

[0037] In this embodiment, the process begins when eye contact with the eyepiece sensor 26 is detected. In this embodiment, the small LCD 24 will hereafter be referred to as the display unit, but the rear LCD 23, or both the rear LCD 23 and the small LCD 24, may also serve as the display unit.

[0038] In step S300, the control unit 203 stores the live view image output from the image sensor 21 in a memory (not shown). Here, the live view image is captured using exposure conditions automatically set and calculated in P mode (program auto). The shooting parameters, including these exposure conditions, used when capturing the live view image are also saved in memory, linked to the live view image.

[0039] In step S301, the control unit 203 starts an adjustment mode for the target image in response to a predetermined user operation on the imaging device 1000.

[0040] In step S302, the control unit 203 receives a request from the user regarding the shooting settings of the imaging device 1000 as voice data. Specifically, when the user speaks the request, the microphone 27 (request acquisition means) acquires the voice, converts it into voice data, and transmits it to the prompt generation unit 205. The prompt generation unit 205 performs character recognition on the voice data acquired from the microphone 27 to generate (convert) a prompt and inputs it to the generation unit 204.

[0041] The prompt generation method in the prompt generation unit 205 will be explained below using Figure 5.

[0042] In this embodiment, the prompt generation unit 205 interprets the user's sensibilities and emotions from the user's voice information and outputs those expressions as prompts. As shown in Figure 5, the prompt generation unit 205 comprises a voice acquisition unit 205a, an emotion interpretation unit 205b, and an emotion expression conversion unit 205c, and uses the functions of these three units to output these prompts. This allows the user to easily communicate their requests regarding the shooting scene using expressions based on their own sensibilities, and the generation unit 204 can then suggest the optimal shooting settings based on those sensibilities.

[0043] The voice acquisition unit 205a acquires audio data of the user's voice from the microphone 27, converts it into text data using a speech recognition algorithm, and sends it to the emotion interpretation unit 205b. In this embodiment, the voice acquisition unit 205a implements a cloud-based speech recognition API such as Google Speech-to-Text or Microsoft Azure Speech Service to convert the acquired audio data into text data. Specifically, when the voice acquisition unit 205a acquires audio data of the user's voice using the microphone 27, it sends the acquired audio data to the API, and obtains text data generated by the API converting the audio data.

[0044] The emotion interpretation unit 205b interprets the user's emotions and sensibilities from the text data transmitted from the voice acquisition unit 205a and transmits the interpretation result to the emotion expression conversion unit 205c. Specifically, the emotion interpretation unit 205b performs natural language processing on the acquired text data to extract keywords related to emotions and sensibilities from the user's statements, or to convert the statements into keywords related to emotions and sensibilities.

[0045] In natural language processing, first, text analysis techniques are used to understand the context and dependencies of text data, and then sentiment analysis techniques are used to extract the user's emotions from the text data.

[0046] Here, text analysis techniques refer to techniques that perform morphological analysis, dependency structure analysis, and contextual analysis.

[0047] Morphological analysis involves dividing text data into words and identifying the part of speech of each divided word. The sentiment interpretation unit 205b performs morphological analysis by implementing a morphological analysis engine such as MeCab or Kuromoji.

[0048] Dependency structure analysis involves analyzing the dependencies between morphologically analyzed words and understanding the sentence structure based on the analysis results. The sentiment interpretation unit 205b performs dependency structure analysis by implementing dependency structure analysis engines such as spaCy and Stanford Parser.

[0049] Contextual analysis uses a pre-trained model to understand the meaning of the entire sentence and to interpret the meaning of specific words and phrases based on their context. The pre-trained model used here may be an existing neural network such as BERT, GPT-3, or GPT-4, or a custom neural network may be constructed. For example, a custom neural network could be constructed to perform contextual analysis based on user-inputted comments about photographs taken by the user, analyzing (learning) the brightness, composition, depth of field, focal length, white balance, etc., of the subject in the user's photographs.

[0050] Furthermore, sentiment analysis technology is a technique for extracting user emotions from text data. This technology makes it possible to generate prompts based on the user's emotions. The emotion interpretation unit 205b may analyze the user's emotions corresponding to specific words or phrases in a rule-based manner by implementing an emotion dictionary such as SentiWordNet. Alternatively, it may analyze the user's emotions corresponding to specific words or phrases by implementing a pre-trained model that has been trained (machine learning) using a dataset with emotion labels in advance. Examples of pre-trained models that can be implemented here include support vector machines (SVM), random forests, and deep learning models (LSTM, BERT).

[0051] Furthermore, any technology capable of extracting user emotions from sources other than text data may be used instead of (or in combination with) the above-mentioned emotion analysis technology. For example, a voice feature extraction technology may be used that extracts features such as tone, pitch, and speed from the voice data acquired by microphone 27, and uses these features to extract the user's emotions. In this case, the emotion interpretation unit 205b extracts voice features by implementing an acoustic feature extraction library (e.g., OpenSMILE or Librosa). Here, voice feature extraction technology is a technology that performs voice data preprocessing, feature extraction, and feature analysis processing. Voice data preprocessing refers to processes such as noise reduction and normalization, and feature extraction refers to the process of extracting features such as tone, pitch, and speed from the preprocessed voice data. Feature analysis refers to the process of analyzing the features extracted in the feature extraction process and estimating emotions.

[0052] The emotion expression conversion unit 205c converts the interpretation result transmitted from the emotion interpretation unit 205b into an ideal expression for the shooting scene, generates a protocol, and transmits it to the generation unit 204. For example, if the user gives a voice instruction via the microphone 27 saying "I want an emotional (sentimental) expression," the emotion expression conversion unit 205c converts this instruction into specific prompts such as "depict with soft tones," "depict the subject softly," and "a brighter scene."

[0053] Returning to Figure 3, in step S303, the generation unit 204 (first generation means) generates a proposed shooting image and shooting parameters based on the live view image held in memory in step S300 and the prompt input from the prompt generation unit 205 in step S302. As will be described in detail later, here, a corrected proposed image that maximizes the shooting effect is generated using a machine learning algorithm.

[0054] In step S304, the control unit 203 displays the proposed image generated by the generation unit 204 in step S303 on the display unit.

[0055] In step S305, the control unit 203 determines whether or not the user has instructed a change in the settings (shooting effect value x) related to the proposed image to be shot. Details of how this setting change instruction is given will be described later, but it is performed by user operation using the operation unit 29 or the electronic dial 28. If there is an instruction to change the settings related to the proposed image to be shot (YES in step S305), the process proceeds to step S306; otherwise, the process proceeds to step S307.

[0056] In step S306, the generation unit 204 (second generation means) generates a proposed shooting image based on the instruction to change the settings related to the proposed shooting image using a machine learning algorithm. Subsequently, the control unit 203 displays the proposed shooting image on the display unit.

[0057] In step S307, the control unit 203 determines whether the shooting parameter settings have been completed. Specifically, when the suggested shooting image displayed on the display unit matches the user's desired shooting image and the user gives the adjustment completion instruction described later, the control unit 203 determines that the shooting parameter settings have been completed. If, as a result of this determination, the shooting parameter settings have been completed (YES in step S307), no further setting changes are necessary, and the process proceeds to step S308. On the other hand, if the suggested shooting image displayed on the display unit differs from the user's desired shooting image and the adjustment completion instruction is not given, the shooting parameter settings have not been completed (YES in step S307), and the process returns to step S305. In this case, the user gives the instruction to change the settings related to the suggested shooting image again.

[0058] In step S308, the shooting parameters for capturing sample images tailored to the user's requirements are calculated. The calculation method will be described later.

[0059] In step S309, once the shooting preparation is complete, which involves setting the shooting parameters using the shooting parameters calculated in step S308, this process terminates.

[0060] In this process, steps S300 to S307 are repeated to reflect the user's preferences, and the process of updating the proposed shooting image generated by the generation unit 204 is repeated, thereby matching the user's preferences. This makes it possible to prepare for shooting using shooting parameters based on the proposed shooting image that reflects the user's preferences.

[0061] Traditionally, users had to manually adjust multiple shooting parameters to achieve their preferred shooting style, resulting in a very complex adjustment process. However, with this invention, users can easily and intuitively set shooting parameters to suit their preferences.

[0062] <How to display the shooting draft image in live view> Figure 4 is a diagram illustrating the proposed shooting image displayed and updated on the display unit in steps S304 and S306 of Figure 3, and the method for issuing setting change instructions in step S305 of Figure 3. Below, the proposed shooting image that is updated each time a setting change instruction is issued on the display unit, and the shooting parameters that are determined and displayed when an adjustment completion instruction is issued, will be explained using Figure 4.

[0063] As an example, Figure 4 shows the shooting parameter settings screen when a user is about to take a portrait in a flower field. It shows the user checking the live view image through the viewfinder unit 206.

[0064] The live view images shown along the time axis in Figure 4 represent the content displayed on the display unit as seen by the user via the viewfinder unit 206. By performing the shooting parameter setting change process shown in Figure 3, the shooting sample images displayed on the display unit are updated and displayed in chronological order.

[0065] At timing t0, a live view image is captured and displayed on the display unit using shooting parameters determined according to the exposure conditions automatically set and calculated by the imaging device 1000 in P mode (program auto).

[0066] At timing t1, the adjustment mode for the proposed image is started in response to a predetermined user operation on the imaging device 1000. At this time, the display unit continues to display the live view image as a proposed image 405 with zero shooting effect (i.e., shooting effect value x is 0). Furthermore, the display unit (adjustment screen display means) displays an adjustment bar 401 below the image for adjusting the shooting effect using a setting adjustment marker 402. The left end 403 of the adjustment bar 401 indicates the faithful setting (i.e., shooting effect value x is 0), and the right end 404 indicates the setting where the shooting effect is maximized (i.e., shooting effect value x is 1). At timing t1, since the current proposed image has no shooting effect, the setting adjustment marker 402 is displayed at the left end 403.

[0067] Furthermore, at timing t1, when the user voices their request for the shooting settings of the imaging device 1000, the microphone 27 converts that voice into audio data. The prompt generation unit 205 then performs character recognition on that audio data to generate text data, and uses that text data to generate a prompt. The generation unit 204 inputs this prompt into a trained model and generates a proposed shooting image 406 that maximizes the shooting effect (i.e., the shooting effect value x is 1) and the shooting parameters for shooting that proposed shooting image as output. Even if the user's request for the shooting settings of the imaging device 1000 at this time is a highly abstract and vague request such as "I want to shoot with a shooting effect that makes people stand out," the generation unit 204 can generate an appropriate proposed shooting image using a pre-trained model.

[0068] At timing t2, the proposed shooting image 406 generated by the generation unit 204 based on the user's voice request at timing t1 is output to the display unit. The position of the setting adjustment marker 402 is moved to the right edge 404 because the shooting effect value x of the proposed shooting image 406 is 1. For example, the generation unit 204 outputs an image as the proposed shooting image 406, which has a shallower background depth and brighter overall subject compared to the proposed shooting image 405 at timing t0.

[0069] Next, at timing t3, the user moves the setting adjustment marker 402 (adjustment means) with respect to the proposed image 406 using the electronic dial 28 or the control unit 29 within the range from the left end 403 to the right end 404 of the adjustment bar 401 (first user operation). At this time, the closer the user moves the setting adjustment marker 402 to the left end 403, the closer the proposed image can be to the live view image (proposed image 405) taken in P mode. In other words, the proposed image can be adjusted to an image with a weaker shooting effect compared to the proposed image 405.

[0070] If the user feels that the shooting effect of the proposed shooting image 406 displayed on the display unit at timing t2 is too strong, they can adjust the effect to be weaker by sliding the setting adjustment marker 402 to the left at timing t3. In this way, the display unit (image display means) displays at least two proposed shooting images, 405 and 406, interchangeably depending on the position of the setting adjustment marker 402.

[0071] Furthermore, the update speed of the proposed image displayed on the display unit when the setting adjustment marker 402 moves does not have to be constant. For example, the update speed of the proposed image displayed on the display unit may be changed to match the speed at which the user moves the setting adjustment marker 402 (operation speed). That is, if the user moves the setting adjustment marker 402 quickly (operation speed is above a threshold), it is assumed that the user intends to significantly change the shooting effect of the proposed image, so the update frequency of the proposed image should be slowed down. On the other hand, if the user is making fine adjustments to the setting adjustment marker 402 (operation speed is below a threshold), it is assumed that the user intends to finely adjust the shooting effect of the proposed image, so the update speed of the proposed image displayed on the display unit should be increased. By changing the update speed of the proposed image in this way, it is possible to reduce the power consumption of the imaging device 1000 and reduce the burden on the user to check the image.

[0072] At timing t4, the generation unit 204 acquires the shooting effect value corresponding to the adjusted position of the setting adjustment marker 402, inputs the acquired shooting effect value into the trained model, and performs inference again to generate a proposed shooting image 407 with adjusted shooting effects. After that, this generated proposed shooting image 407 is displayed on the display unit. At this time, the generation unit does not need to generate the shooting parameters. It should be assumed that the trained model used here has undergone additional training using proposed shooting image 405 (shooting effect value x=0) and proposed shooting image 406 (shooting effect value x=1).

[0073] Shooting proposal image 407 is a shooting proposal image that reduces the shooting effect requested by the user compared to shooting proposal image 406. For example, it is an image in which the brightness of the subject is reduced compared to shooting proposal image 406.

[0074] In this way, the user moves the setting adjustment marker 402 until a draft image with the desired shooting effect is generated. Then, at timing t5, when a draft image with a satisfactory shooting effect is displayed on the display unit, the user presses the draft image confirmation button (not shown) (second user operation). In response to this confirmation button being pressed, an adjustment completion instruction is sent to the generation unit 204.

[0075] At timing t6, the generation unit 204 receives an instruction to complete the adjustment, obtains an image capture effect value corresponding to the current position of the setting adjustment marker 402, inputs the obtained image capture effect value into the trained model, and performs inference again to generate image capture parameters with adjusted image capture effects. The control unit 203 (image capture preparation means) uses these generated image capture parameters to perform image capture preparation. At this time, as shown on screen 408, an image similar to the image capture proposal image displayed at timing t5 and a list of image capture parameters generated at timing t6 may be displayed.

[0076] Thus, in this embodiment, the user can select a shooting image with a satisfactory shooting effect simply by referring to the shooting image generated by the generation unit 204 and displayed and updated on the display unit, adjusting the position of the setting adjustment marker 402, and pressing the confirm button. Furthermore, upon pressing the confirm button, the imaging device 1000 sets the shooting parameters to obtain the same shooting effect as the selected shooting image, allowing for shooting with the same satisfactory effect on the actual subject.

[0077] Next, we will describe the configuration of the trained model that generates the proposed image and shooting parameters, which is provided in the generation unit 204, and the method of training it.

[0078] Figure 5 shows an example configuration of the generator 204 equipped with one trained model.

[0079] As shown in Figure 5, the camera type and lens type of the imaging device 1000 currently used by the user, prompts representing the user's preferences, the shooting effect value x, the live view image, and the shooting parameters associated with the live view image become the input data for the trained model. Here, the initial value of the shooting effect value x is the maximum value of 1, and when the position of the setting adjustment marker 402 in Figure 4 is moved by user operation, the value of the shooting effect value x changes according to that movement.

[0080] Furthermore, multiple pre-trained models may be prepared for each camera type and lens type. In this case, prompts representing user preferences and live view images are input data to the single pre-trained model used.

[0081] As shown in Figure 5, one trained model in the generation unit 204 outputs a proposed image and shooting parameters as output data. In such a trained model, the proposed image and shooting parameters are closely related to each other during generation, so the shooting result obtained using the generated shooting parameters is closer to the proposed image. On the other hand, when multiple tasks are given to a single trained model, the model size generally increases and processing time tends to increase.

[0082] Figure 6 shows an example configuration of the generator 204, which has two trained models.

[0083] In Figure 6, when the prompt, live view image, shooting parameters associated with the live view image, and shooting effect values ​​are first input data (first input data), a proposed shooting image is output as output data. Next, the proposed shooting image output from trained model A (second input data) and the camera type and lens type of the imaging device 1000 currently used by the user are input to trained model B (second trained model), and the shooting parameters are output as output data. In this configuration, trained model A is a model that outputs a proposed shooting image with shooting effects applied to the input live view image, and trained model B is a model that searches for shooting parameters to realize the proposed shooting image. Therefore, the tasks are divided and simplified compared to the configuration in Figure 5, and the processing accuracy for each task can be improved.

[0084] In this embodiment, a generation unit 204 is provided within the imaging device 1000, and since it is necessary to reduce the processing load, it is preferable to use the configuration example in Figure 6 rather than the configuration example in Figure 5.

[0085] Next, we will explain the learning method of this embodiment.

[0086] Since the generation unit 204 in this embodiment ultimately derives the shooting parameters, it is necessary to calculate shooting parameters that take into account the characteristics of the equipment. Furthermore, user requests must be reflected as prompts and input data in the proposed shooting image and the shooting parameters based on it.

[0087] Therefore, the training model that forms the basis of the pre-trained model used in the generation unit 204 is a model that takes camera type and lens type information as input data and is capable of performing inference processing specific to each camera type and lens type. In addition, it is necessary to generate suggested shooting images that match the user's preferences as prompts. For this reason, separate from the training of suggested shooting images and shooting parameters, the pre-trained model of the prompt generation unit 205 is also trained in advance to perform syntactic analysis of prompts and determine information related to shooting parameters.

[0088] First, the base learning model is repeatedly trained using a large number of captured images associated with shooting parameters as training data. This process builds a pre-trained model that, when a live view image is input, outputs generally preferred shooting parameters and suggested images. Furthermore, additional training is performed using shooting parameters associated with prompts as training data, and using suggested images output from the pre-trained model as input data. During this training, captured images that match the user's preferences are registered in the generation unit 204 beforehand. This allows the pre-trained model used by the generation unit 204 to match the user's preferences. The registered captured images can be prepared by the user or presented by the generation unit 204. In this embodiment, the captured images presented by the generation unit 204 are registered in the generation unit 204 when selected by the photographer.

[0089] Furthermore, the generation unit 204 presents the captured images in stages. In this embodiment, when the user selects the first captured image to be registered in the generation unit 204, the generation unit 204 presents images similar to the selected image as options for the next image to be registered, in order to narrow down the user's preferences, and prompts the user to make a selection. This staged presentation may continue until the generation unit 204 has narrowed down the user's preferences, or it may continue until the user instructs to end the registration process.

[0090] By collecting the images to be registered in this way and training the generation unit 204, it becomes possible to generate suggested images that are closer to the user's preferences compared to the suggested images generated by the initially trained model.

[0091] In this embodiment, when the user is asked to select the registration information to be used for learning, a photograph was presented. However, this is not the only option; the user may also be asked to make a selection using a text-based question format.

[0092] However, the explanation so far only covers the initial setup. In reality, many shots will be taken from this initial state, and the results of these shots will also be used to further train the pre-trained model in the generation unit 204. The user repeatedly adjusts the proposed images generated by the generation unit 204 and inputs voice requests regarding the images to the microphone 27. The proposed images and shooting parameters adopted by the user at this time are linked to their prompts and used for further training in the generation unit 204. In this way, it is possible to generate a trained model that takes the user's preferences into further consideration.

[0093] Furthermore, a time-series inference flag may be added as input information, allowing the user to choose whether or not to learn from past user-generated image suggestions. This allows for situations where the user does not want the generation unit 204 to perform time-series inference. In this embodiment, the user can pre-instruct the generation unit 204 whether or not to use the most recently adjusted image suggestions for further training of the trained model. However, this is not limited to this embodiment; the trained model of the generation unit 204 may be configured to make its own decision on whether or not to perform time-series inference, or it may be a trained model that always performs time-series inference.

[0094] Additionally, users may be allowed to select images from the proposed images being prepared for further training.

[0095] As described above, the imaging device 1000 generates a proposed image and shooting parameters to realize it during shooting from the live view image and prompts before shooting, displays the proposed image, and allows the user to adjust the shooting effect of the proposed image. This provides a shooting assist function that is suitable for the user.

[0096] Furthermore, in this embodiment, when selecting or deciding on a shooting option image by user operation using the electronic dial 28 or the control unit 29, the corrected option image may be displayed overlaid on the live view image. This allows the user to switch between shooting option images while comparing them with the live view image, enabling them to select the shooting option image that best suits their preferences.

[0097] In this embodiment, the start timing of the process shown in Figure 3 is set to the time when an eye is detected on the eyepiece sensor 26, but this is not limited to this. For example, the process shown in Figure 3 may be continuously executed from the time when the acquisition of the live view image begins.

[0098] In summary, this embodiment allows even users unfamiliar with photography to easily take photos that match their intentions. Furthermore, since shooting parameters are automatically generated by a machine learning algorithm, optimal settings are suggested for each shooting scene, improving the quality of the resulting images.

[0099] Although Example 1 has been described in detail above, the present invention is not limited to a specific embodiment, and various modifications and changes are possible within the scope described in the claims. Furthermore, it is possible to combine all or more of the components of this embodiment.

[0100] For example, in this embodiment, an example was described in which a suggested shooting image is displayed and the user operates the setting adjustment marker 402. However, it is also possible for the user to move the position of the setting adjustment marker 402 left or right without displaying the suggested shooting image, and for the generation unit 204 to generate shooting parameters with adjusted shooting effects according to the position after the movement.

[0101] Regardless of whether the proposed shooting image is displayed to the user through the display unit, the generation unit 204 generates shooting parameters related to the proposed shooting image, so it is not necessarily required to present the proposed shooting image to the user.

[0102] (Second example) A second embodiment of the present invention will now be described. The imaging device 1001 according to this embodiment differs from the imaging device 1000 according to the first embodiment in that it does not have a generation unit 204. Furthermore, the imaging device 1001 differs in that the generation process performed by the generation unit 204 is performed by an external generation device 1002 connected to the imaging device 1001.

[0103] In other words, this embodiment is performed by an imaging system consisting of an imaging device 1001 and a generating device 1002.

[0104] In this embodiment, the same numbering will be used for configurations and steps similar to those in the first embodiment, and redundant explanations will be omitted.

[0105] In the first embodiment, the generation unit 204 preferably includes two trained models (Figure 6) as described above. This is because the processing in the generation unit 204 has a greater processing load than general image processing performed by the imaging device. Therefore, in this embodiment, the processing in the generation unit 204 is performed in a generation device 1002 (Figure 10), which is a separate device from the imaging device 1001.

[0106] Figure 9 is a block diagram showing the hardware configuration of the imaging device 1001 in this embodiment. Hereafter, components of the imaging device 1001 that are the same as those in Figures 1A, 1B, and 2 will be denoted by the same reference numerals, and redundant explanations will be omitted.

[0107] In Figure 9, the imaging device 1001 does not have a generation unit 204, but instead has a wireless transceiver unit 901. The wireless transceiver unit 901 is capable of sending and receiving data with external devices, and is used to communicate data with the generation device 1002 (Figure 10), which will be described later.

[0108] Figure 10 shows the imaging system in this embodiment.

[0109] The imaging system shown in Figure 10 consists of an imaging device 1001 and a generation device 1002, which are connected in a communicative manner, as described in Figure 9. The generation device 1002 is a device capable of performing the same processing as the generation unit 204 shown in the first embodiment, and includes a generation unit and a wireless transceiver unit, which are not shown.

[0110] In the imaging system shown in Figure 10, the generation device 1002 is responsible for the processing performed by the generation unit 204 in the first embodiment. Therefore, the live view image, audio data, prompts, shooting effect values, shooting parameters associated with the live view image, and time-series inference flags are transmitted from the imaging device 1001 to the generation device 1002 as communication data via the wireless transceiver unit 901. On the other hand, the generation device 1002 performs inference processing using the communication data transmitted from the imaging device 1001 as input data in a generation unit (not shown) of the generation device 1002, generating a proposed shooting image and shooting parameters, which are then transmitted to the imaging device 1001. Other processing is the same as in the first embodiment.

[0111] Next, using the flowcharts in Figures 7 and 8, the processing of the imaging system realized by the interaction between the imaging device 1001 and the generation device 1002 in this embodiment will be explained. Note that the same reference numerals are used for steps similar to those in the first embodiment, and redundant explanations are omitted.

[0112] The process shown in Figure 7 is achieved by the CPU 201 sending commands to the control unit 203 to control various parts of the imaging device 1001, similar to the process shown in Figure 3. In this embodiment, the process is initiated when eye contact with the eyepiece sensor 26 is detected.

[0113] First, steps S300 to S302 are executed, and then the process proceeds to step S701.

[0114] In step S701, the control unit 203 transmits communication data to the generation device 1002 via the wireless transceiver unit 901 (first data transceiver unit). The communication data (first communication data) transmitted to the generation device 1002 includes voice data of requests for the suggested images to be captured, voice prompts generated from the voice data by the prompt generation unit 205, and live view images.

[0115] In step S702, the control unit 203 determines whether it has received the proposed image and shooting parameters as communication data from the generation device 1002 via the wireless transceiver unit 901 (first data transceiver unit). This communication data is transmitted to the imaging device 1001 in step S802, as shown in Figure 8, which will be described later. If the communication data from the generation device 1002 has been received (YES in step S702), step S304 is executed, and the process proceeds to step S305. On the other hand, if the communication data from the generation device 1002 has not been received (NO in step S702), the process returns to step S701.

[0116] In step S305, the control unit 203 determines whether or not the user has instructed a change in the settings (shooting effect value x) related to the proposed image to be shot, similar to the first embodiment. However, if there is an instruction to change the settings related to the proposed image to be shot (YES in step S305), the process proceeds to step S703; otherwise, the process proceeds to step S307.

[0117] In step S703, the control unit 203 generates an additional prompt including a shooting effect value x determined according to the position of the setting adjustment marker 402. This ensures that, as in the first embodiment, if the user changes the settings (shooting effect) for the proposed image in step S305, the shooting effect of the proposed image is changed as intended by the user. Then, returning to step S701, the additional prompt is sent to the generation device 1002 via the wireless transceiver 901 (second data transceiver), and the proposed image with the adjusted shooting effect and shooting parameters are obtained from the generation device 1002. The shooting effect value x included in the additional prompt is information indicating the magnitude of the shooting effect compared to the proposed image (live view image) displayed as the faithful setting.

[0118] Next, we will explain the processing on the generation device 1002 side using Figure 8.

[0119] The process shown in Figure 8 is realized by a CPU (not shown) in the generation device 1002 controlling various parts of the generation device 1002.

[0120] In step S801, it is determined whether communication data from the imaging device 1001 has been received by the wireless transceiver unit (not shown) of the generation device 1002. The communication data here is the data transmitted from the imaging device 1001 in step S701 described above. If communication data from the imaging device 1001 has been received (YES in step S801), the process proceeds to step S303 to generate a draft image and shooting parameters, and then proceeds to step S802. On the other hand, if communication data from the imaging device 1001 has not been received (NO in step S801), the process returns to step S801.

[0121] In step S802, the proposed image and shooting parameters generated in step S303 are transmitted as communication data to the imaging device 1001, and this process ends. The process on the generation device 1002 side shown in Figure 8 is repeatedly executed until the generation device 1002 shuts down.

[0122] As described above, in this embodiment, the computationally intensive processing performed by the generation unit 204 in the first embodiment is performed by a separate device, the generation device 1002, instead of the imaging device 1001. This makes it possible to reduce the processing load on the imaging device 1001. Furthermore, by providing the generation device 1002 independently as the device that performs the computationally intensive processing, it becomes possible to easily update the trained model used in the processing, and thus improve the performance of the processing in a simple manner.

[0123] Even in a configuration where the generation unit 204 performs the same processing, as in the imaging device 1001 of the first embodiment, updating the trained model by downloading the trained model via a network is within the scope of this embodiment.

[0124] (Other embodiments) The present invention can also be realized by supplying a program that implements one or more of the functions of the above embodiments to a system or device via a network or storage medium, and by having one or more processors in the computer of that system or device read and execute the program. It can also be realized by a circuit (e.g., an ASIC) that implements one or more functions.

[0125] Furthermore, the imaging device according to the present invention is not limited to a device having only an imaging function, such as the imaging devices 1000 and 1001 according to the above embodiment, but may also be an information processing device that is equipped with functions other than imaging, such as a smartphone.

[0126] Although preferred embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and various modifications and changes are possible within the scope of its gist.

[0127] This embodiment includes the following configurations, methods, and programs. (Configuration 1) An imaging device comprising: a shooting means; a request acquisition means for acquiring a request for a shooting effect from a user; a conversion means for converting the request for the shooting effect into a prompt; a first generation means for generating output data, which includes a live view image obtained by the shooting means, shooting parameters associated with the live view image, and the prompt, using a machine learning algorithm to generate shooting parameters and a shooting suggestion image that maximize the shooting effect; an image display means for switching between displaying at least the live view image with zero shooting effect and the shooting suggestion image with maximum shooting effect; an adjustment means for adjusting the shooting effect within a range from zero to maximum in response to a first user operation; a second generation means for generating shooting parameters that result in the adjusted shooting effect using the machine learning algorithm; and a shooting preparation means for performing shooting preparation using the adjusted shooting parameters when a second user operation is received. (Configuration 2) The imaging device according to Configuration 1, further comprising an adjustment bar for adjusting the shooting effect from zero to the maximum range, and an adjustment screen display means for displaying a marker indicating the current shooting effect above the adjustment bar, wherein the first user operation is an operation to move the position of the marker above the adjustment bar. (Configuration 3) The imaging apparatus according to Configuration 1 or 2, characterized in that the machine learning algorithm uses a trained model obtained by repeatedly training with captured images associated with the shooting parameters as training material in its initial state. (Configuration 4) The imaging device according to Configuration 3, characterized in that the trained model is constructed according to the camera type and lens type of the imaging device. (Configuration 5) The imaging device according to Configuration 3 or 4, characterized in that the input data includes an imaging effect value indicating the magnitude of the imaging effect, and the initial value of the imaging effect value is the maximum value. (Configuration 6) The imaging device according to Configuration 5, characterized in that the trained model consists of one trained model that, when input data including the live view image, shooting parameters associated with the live view image, the prompt, and the shooting effect value is input, outputs the shooting parameters of the shooting effect indicated by the shooting effect value and a suggested shooting image as output data. (Configuration 7) The imaging device according to Configuration 5, characterized in that the trained model comprises a first trained model that outputs a proposed image of the shooting effect indicated by the shooting effect value as output data when first input data including the live view image, shooting parameters associated with the live view image, the prompt, and the shooting effect value is input, and a second trained model that outputs the shooting parameters of the shooting effect indicated by the shooting effect value as output data when second input data including the proposed image output from the first trained model is input. (Configuration 8) The imaging device according to any one of Configurations 1 to 7, characterized in that the adjustment means adjusts a plurality of camera setting items that affect the shooting effect as the shooting parameters. (Configuration 9) The imaging device according to Configuration 8, characterized in that the camera setting items include at least one of the following: shutter speed, aperture value, ISO sensitivity, white balance, exposure, image contrast, saturation, sharpness, lighting correction, tone priority, shooting style, noise reduction, color correction, subject tracking setting, focus setting, and image processing setting. (Configuration 10) The imaging apparatus according to any one of Configurations 1 to 9, characterized in that the second generation means further outputs a proposed image obtained by obtaining the adjusted shooting effect using the machine learning algorithm each time the shooting effect is adjusted by the adjustment means, and the image display means updates the display to the proposed image obtained by obtaining the adjusted shooting effect. (Configuration 11) The imaging apparatus according to Configuration 10, characterized in that the image display means changes the update speed of the display on the shooting proposal image from which the adjusted shooting effect is obtained, according to the operation speed of the first user operation. (Configuration 12) The imaging apparatus according to Configuration 11, characterized in that the update speed is increased when the operating speed of the first user operation is less than a threshold, and the update speed is decreased when the operating speed of the first user operation is equal to or greater than the threshold. (Configuration 13) The imaging apparatus according to any one of Configurations 1 to 12, characterized in that the image display means is at least one of a rear panel and an electronic viewfinder (EVF). (Configuration 14) The imaging device according to any one of Configurations 1 to 13, characterized in that the request acquisition means acquires the request as voice data obtained by acquiring the voice spoken by the user and recognizing the characters. (Configuration 15) An imaging device connected to a generation device, the imaging device comprising: an imaging means; a request acquisition means for acquiring a request for an imaging effect from a user; a conversion means for converting the request for an imaging effect into a prompt; a first data transmission / reception unit that transmits a first communication data to the generation device, including a live view image obtained by the imaging means, imaging parameters associated with the live view image, and the prompt, and receives from the generation device the imaging parameters and suggested image that maximize the imaging effect, which are generated from the first communication data using a machine learning algorithm; and at least the live An imaging device comprising: an image display means that can switch between displaying a view image and a shooting plan image that maximizes the shooting effect; an adjustment means that adjusts the shooting effect within a range from zero to maximum in response to a first user operation; a second data transmission / reception unit that transmits a shooting effect value indicating the magnitude of the adjusted shooting effect to the generation device and receives from the generation device the shooting parameters and shooting plan image of the adjusted shooting effect generated from the shooting effect value using the machine learning algorithm; and a shooting preparation means that, when a second user operation occurs, performs shooting preparation using the shooting parameters of the adjusted shooting effect. (Method 1) A control method characterized by comprising: a shooting step; a request acquisition step for acquiring a request for a shooting effect from a user; a conversion step for converting the request for a shooting effect into a prompt; a first generation step for generating output data, which includes a live view image obtained in the shooting step, shooting parameters associated with the live view image, and the prompt, using a machine learning algorithm to generate shooting parameters and a suggested shooting image that maximize the shooting effect; an image display step for switching between displaying at least the live view image with zero shooting effect and the suggested shooting image with maximum shooting effect; an adjustment step for adjusting the shooting effect within a range from zero to maximum in response to a first user operation; a second generation step for generating shooting parameters that result in the adjusted shooting effect using the machine learning algorithm; and a shooting preparation step for executing shooting preparation using the adjusted shooting parameters when a second user operation occurs. (Method 2) A method for controlling an imaging device connected to a generation device, wherein the imaging device includes a shooting step, a request acquisition step for acquiring a request for a shooting effect from a user, a conversion step for converting the request for the shooting effect into a prompt, a first data transmission / reception unit that transmits first communication data including a live view image obtained in the shooting step, shooting parameters associated with the live view image, and the prompt to the generation device, and receives from the generation device the shooting parameters and shooting suggestion image that maximize the shooting effect, generated from the first communication data using a machine learning algorithm, and at least the shooting effect is zero A control method characterized by comprising: an image display step that switches between displaying an Eveview image and a shooting plan image that maximizes the shooting effect; an adjustment step that adjusts the shooting effect within a range from zero to maximum in response to a first user operation; a second data transmission / reception unit that transmits a shooting effect value indicating the magnitude of the adjusted shooting effect to the generation device and receives from the generation device the shooting parameters of the adjusted shooting effect and the shooting plan image generated from the shooting effect value using the machine learning algorithm; and a shooting preparation step that, when a second user operation occurs, performs shooting preparation using the shooting parameters of the adjusted shooting effect. (Program 1) A program for causing a computer to function as one of the means of an imaging device described in any one of configurations 1 to 14. (Program 2) A program for causing the computer to function as each of the means of the imaging apparatus described in Configuration 15. [Explanation of Symbols]

[0128] 100 imaging optical system 11 aperture 12 Image stabilization lens group 13 Focusing lens group 200 Camera body 20 Electrical Circuits 21 Image sensor 22 Mechanical shutter 23 Rear LCD 24 Small LCD 25 Optical system 26 Eyepiece Sensor 27 Mike 28 Electronic Dial 29 Control section 201 CPU 202 Image Processing Unit 203 Control Unit 204 Generation part 205 Prompt generation unit 901 Wireless Transceiver Unit 1000,1001 Imaging device 1002 Generator

Claims

1. Photography methods, A means for obtaining requests for shooting effects from users, A conversion means that converts the aforementioned shooting effect requests into prompts, A first generation means generates output data consisting of shooting parameters and a suggested shooting image that maximize the shooting effect, using a machine learning algorithm from input data including a live view image obtained by the shooting means, shooting parameters associated with the live view image, and the prompt. At a minimum, an image display means that can switch between displaying the live view image in which the shooting effect is zero and the proposed shooting image in which the shooting effect is maximized, An adjustment means for adjusting the shooting effect within a range from zero to maximum in response to a first user operation, A second generation means for generating the adjusted shooting parameters using the machine learning algorithm, When a second user operation occurs, a shooting preparation means executes shooting preparation using the adjusted shooting parameters of the shooting effect, An imaging device characterized by comprising:

2. The system further includes an adjustment bar for adjusting the aforementioned shooting effect from zero to the maximum range, and an adjustment screen display means for displaying a marker above the adjustment bar indicating the current shooting effect. The imaging apparatus according to claim 1, characterized in that the first user operation is an operation to move the position of the marker on the adjustment bar.

3. The imaging apparatus according to claim 1, characterized in that the machine learning algorithm uses a trained model obtained by repeatedly training with captured images associated with shooting parameters as training material in its initial state.

4. The imaging device according to claim 3, characterized in that the trained model is constructed according to the camera type and lens type of the imaging device.

5. The input data includes a shooting effect value indicating the magnitude of the shooting effect. The imaging apparatus according to claim 3, characterized in that the initial value of the aforementioned imaging effect value is the maximum value.

6. The imaging device according to claim 5, characterized in that the trained model consists of one trained model that, when input data including the live view image, shooting parameters associated with the live view image, the prompt, and the shooting effect value is input, outputs the shooting parameters of the shooting effect indicated by the shooting effect value and a suggested shooting image as output data.

7. The imaging apparatus according to claim 5, characterized in that the trained model comprises a first trained model that, when first input data including the live view image, shooting parameters associated with the live view image, the prompt, and the shooting effect value is input, outputs a proposed shooting image of the shooting effect indicated by the shooting effect value as output data, and a second trained model that, when second input data including the proposed shooting image output from the first trained model is input, outputs the shooting parameters of the shooting effect indicated by the shooting effect value as output data.

8. The imaging apparatus according to claim 1, characterized in that the adjustment means adjusts a plurality of camera setting items that affect the shooting effect as the shooting parameters.

9. The imaging device according to claim 8, characterized in that the camera setting items include at least one of the following: shutter speed, aperture value, ISO sensitivity, white balance, exposure, image contrast, saturation, sharpness, lighting correction, tone priority, shooting style, noise reduction, color correction, subject tracking setting, focus setting, and image processing setting.

10. The second generation means further outputs a proposed image for obtaining the adjusted shooting effect using the machine learning algorithm each time the shooting effect is adjusted by the adjustment means. The imaging apparatus according to claim 1, characterized in that the image display means updates the display to a proposed image from which the adjusted shooting effect can be obtained.

11. The imaging apparatus according to claim 10, characterized in that the image display means changes the update speed of the display on the shooting proposal image from which the adjusted shooting effect is obtained, according to the operation speed of the first user operation.

12. The imaging apparatus according to claim 11, characterized in that the update speed is increased when the operating speed of the first user operation is less than a threshold, and the update speed is decreased when the operating speed of the first user operation is equal to or greater than the threshold.

13. The imaging apparatus according to claim 1, characterized in that the image display means is at least one of a rear panel and an electronic viewfinder (EVF).

14. The imaging device according to claim 1, characterized in that the request acquisition means acquires the request as voice data obtained by acquiring the voice spoken by the user and recognizing the text.

15. An imaging device connected to a generating device, The imaging device is Photography methods, A means for obtaining requests for shooting effects from users, A conversion means that converts the aforementioned shooting effect requests into prompts, A first data transmission / reception unit transmits first communication data, including a live view image obtained by the shooting means, shooting parameters associated with the live view image, and the prompt, to the generation device, and receives from the generation device the shooting parameters and shooting suggestion image that maximize the shooting effect, which are generated from the first communication data using a machine learning algorithm. At a minimum, an image display means that can switch between displaying the live view image in which the shooting effect is zero and the proposed shooting image in which the shooting effect is maximized, An adjustment means for adjusting the shooting effect within a range from zero to maximum in response to a first user operation, A second data transmission / reception unit transmits the adjusted shooting effect value, which indicates the magnitude of the adjusted shooting effect, to the generation device, and receives the adjusted shooting parameters and proposed shooting image, which are generated from the shooting effect value using the machine learning algorithm, from the generation device. When a second user operation occurs, a shooting preparation means executes shooting preparation using the adjusted shooting parameters of the shooting effect, An imaging device characterized by comprising:

16. Shooting steps, A request acquisition step to obtain requests for shooting effects from users, A conversion step that converts the aforementioned shooting effect requests into prompts, A first generation step involves using a machine learning algorithm to generate output data consisting of the shooting parameters and suggested image that maximize the shooting effect, based on the input data including the live view image obtained in the shooting step, the shooting parameters associated with the live view image, and the prompt. At a minimum, the image display step includes switching between displaying the live view image in which the shooting effect is zero and the proposed shooting image in which the shooting effect is maximized, An adjustment step to adjust the shooting effect within a range from zero to maximum in response to a first user operation, A second generation step involves generating the adjusted shooting parameters using the machine learning algorithm, Upon receiving a second user action, a shooting preparation step is performed, in which shooting preparation is carried out using the shooting parameters of the adjusted shooting effect. A control method characterized by having the following features.

17. A method for controlling an imaging device connected to a generating device, The imaging device is Shooting steps, A request acquisition step to obtain requests for shooting effects from users, A conversion step that converts the aforementioned shooting effect requests into prompts, A first data transmission / reception unit transmits first communication data, including the live view image obtained in the shooting step, the shooting parameters associated with the live view image, and the prompt, to the generation device, and receives from the generation device the shooting parameters and shooting suggestion image that maximize the shooting effect, which are generated from the first communication data using a machine learning algorithm. At a minimum, the image display step includes switching between displaying the live view image in which the shooting effect is zero and the proposed shooting image in which the shooting effect is maximized, An adjustment step to adjust the shooting effect within a range from zero to maximum in response to a first user operation, A second data transmission / reception unit transmits the adjusted shooting effect value, which indicates the magnitude of the adjusted shooting effect, to the generation device, and receives the adjusted shooting parameters and proposed shooting image, which are generated from the shooting effect value using the machine learning algorithm, from the generation device. Upon receiving a second user action, a shooting preparation step is performed, in which shooting preparation is carried out using the shooting parameters of the adjusted shooting effect. A control method characterized by having the following features.

18. A program for causing a computer to function as each of the means of the imaging apparatus described in claim 1.

19. A program for causing a computer to function as each of the means of the imaging apparatus described in claim 15.

Images