Information processing device, information processing method

Abstract

To provide a technology for presenting, in content generated by a generative model, the regions generated by the generative model and / or regions presumed to have been generated by the generative model, in a more intuitive way. [Solution] In content generated by a generative model, the regions generated by the generative model and / or regions that are presumed to have been generated by the generative model are detected based on information related to content generation, and the detected regions are presented.

Description

Technical Field

[0001] The present invention relates to a technology for processing content generated by a generation model.

Background Art

[0002] With the spread of generative AI, an environment is being prepared in which individuals can easily generate large amounts of various types of content (text, images, videos, audio, 3D models, etc.). Generative AI can generate content from scratch or modify and process existing content by inputting a prompt. Also, regarding the modification and processing of existing content, objects in an image or video can be deleted or added. In the case of audio, a specific voice can be deleted or changed. However, when a person other than the creator of the generated content plays the content, it is difficult to determine which part was generated by the generative AI. If there is information such as the instruction coordinates and generated content at the time of generation given to the content, by checking such information, it is possible to confirm what was generated where in the content.

[0003] Patent Document 1 discloses a technique that enables restoring unprocessed image data from processed image data without image quality degradation by configuring the processed image data with a combination of unprocessed image data and processing information representing the processing applied thereto. Since the processing information records what processing was applied to which coordinates, the user can check the details of the processing.

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 disclosed in Patent Document 1, although coordinate information of the modified / processed area in the image is retained, it is difficult to intuitively understand which area of ​​the image has been modified or processed because it is recorded as coordinate values. Also, since the original image remains, there is a concern that things that the creator intended to remove may still be visible. Furthermore, if the generative model used for modification / processing is unavailable, there is a problem in that the generated image cannot be reproduced.

[0006] The present invention provides a technology for presenting, in content generated by a generative model, the regions generated by the generative model and / or regions presumed to have been generated by the generative model, in a way that allows for more intuitive confirmation. [Means for solving the problem]

[0007] One aspect of the present invention is characterized by comprising: a detection means for detecting, based on information related to content generation, regions generated by a generation model and / or regions presumed to have been generated by a generation model in content generated by the generation model; and a presentation means for presenting the regions detected by the detection means. [Effects of the Invention]

[0008] According to the present invention, it is possible to provide a technology for presenting, in content generated by a generative model, the regions generated by the generative model and / or regions that are presumed to have been generated by the generative model in a more intuitive manner. [Brief explanation of the drawing]

[0009] [Figure 1] A block diagram showing an example of a computer device's hardware configuration. [Figure 2] A block diagram showing an example of the functional configuration of the information processing device 100. [Figure 3] A flowchart illustrating the operation of the information processing device 100. [Figure 4] (a) is a figure showing the original still image, and (b) is a figure showing the still image generated by the generative model. [Figure 5] A diagram showing an example of the display of the presentation window 400. [Modes for carrying out the invention]

[0010] The embodiments will be described in detail below with reference to the attached drawings. Note that the following embodiments do not limit the invention to the claims. While the embodiments describe multiple features, not all of these features are essential to the invention, and the features may be combined in any way. Furthermore, in the attached drawings, the same or similar configurations are given the same reference numerals, and redundant descriptions are omitted.

[0011] [First Embodiment] The information processing device according to this embodiment takes content generated by a "generation model that generates new content from input content according to specified conditions" as target content, identifies the region generated by the generation model and / or the region that is presumed to have been generated by the generation model within the target content by analyzing information related to the generation of the target content, and presents the identified region to the user.

[0012] In this embodiment, we describe a case where the target content is content created by partially modifying the original content using a pre-trained generative model. For example, the content is an image in which a part of the original image has been modified using a generative model that has an inpainting function to change a part of the image.

[0013] An example of the functional configuration of the information processing device 100 according to this embodiment is shown in the block diagram in Figure 2. The operation of the information processing device 100 according to this embodiment will be explained according to the flowchart in Figure 3.

[0014] In step S201, the acquisition unit 110 acquires the content generated by the generation model. The method of acquiring the content is not limited to a specific method. For example, the acquisition unit 110 may acquire content stored in non-volatile memory such as a hard disk of the information processing device 100. Alternatively, the acquisition unit 110 may acquire content stored in the memory of an external server device that can communicate with the information processing device 100 via a network. The content may include, for example, text, still images, moving images, audio information, and 3D models. If the acquisition unit 110 acquires multiple types of content (for example, text, moving images, and audio information), the processing from step S202 onward will be performed for each type of content.

[0015] In step S202, the acquisition unit 111 acquires information about the content acquired in step S201. The information about the content is, for example, metadata of the content. For example, if the content is a still image, the metadata of the content is metadata such as Exif (Exchangeable Image File Format).

[0016] In step S203, the analysis unit 112 determines whether the information obtained in step S202 includes information related to content generation (generation information). If the information obtained in step S202 includes generation information, the process proceeds to step S204. If the information obtained in step S202 does not include generation information, the process according to the flowchart in Figure 3 is terminated.

[0017] The generated information includes, for example, information about the generation model (e.g., the name of the generation model, the functions and characteristics of the generation model), "location information representing the location of the area generated by the generation model" specified when the content is generated by the generation model, and "prompts" specified when the content is generated by the generation model.

[0018] In step S204, the analysis unit 112 analyzes the information obtained in step S202, and obtains, as the result of the analysis, the presence or absence of "in the content generated by the generation model, the area generated by the generation model and / or the area estimated to be generated by the generation model", and various information related to the area.

[0019] In this embodiment, the analysis unit 112 includes an area analysis unit 113, a prompt analysis unit 114, and a model analysis unit 115. The area analysis unit 113, the prompt analysis unit 114, and the model analysis unit 115 are functional units that can operate independently, and the analysis unit 112 obtains the result of the analysis by any one of the area analysis unit 113, the prompt analysis unit 114, and the model analysis unit 115.

[0020] The area analysis unit 113 obtains "position information representing the position of the generation area generated by the generation model in the content" included in the information obtained in step S202. Then, the area analysis unit 113 specifies the area corresponding to the position information in the content as the generation area.

[0021] For example, when the content is a still image, the area analysis unit 113 obtains "position information representing the position of the image area generated by the generation model in the still image" included in the information obtained in step S202. Then, the area analysis unit 113 specifies the image area corresponding to the position information in the still image as the generation area.

[0022] Also, for example, when the content is a moving image, the area analysis unit 113 obtains "the identifier of the frame including the image area generated by the generation model in the moving image" and "position information representing the position of the image area generated by the generation model in the frame" included in the information obtained in step S202. Then, the area analysis unit 113 specifies the image area corresponding to the position information in the frame corresponding to the identifier in the moving image as the generation area.

[0023] For example, if the content is audio information, the region analysis unit 113 obtains the "identifier of the frame (time) containing the audio generated by the generation model in the audio information" which is included in the information obtained in step S202. The region analysis unit 113 then identifies the frame corresponding to this identifier in the audio information as the generation region.

[0024] The prompt analysis unit 114 obtains the "prompt specified for content generation" contained in the information acquired in step S202. The prompt analysis unit 114 then analyzes the prompt using existing text mining techniques to interpret the "wording related to location," "wording related to the target object," and "wording related to the execution content such as generation or deletion." Based on the results of this interpretation, the prompt analysis unit 114 estimates the generation area in the content generated by the generation model.

[0025] For example, if the content is a still image and the prompt is "Draw a flower in the lower right corner of the still image," the prompt analysis unit 114 recognizes from the words "lower right" and "flower" that a "flower" object is generated in the lower right area of ​​the still image, and estimates that the area of ​​the "flower" object is highly likely to be the generation area (generation probability).

[0026] For example, if the content is a still image and the prompt is "Remove the person on the left in the still image," the prompt analysis unit 114 recognizes from the words "person on the left" and "remove" that the "person" object is being removed from the left of the group of objects included in the still image, and estimates that the area to the left of the group of objects is likely to be the generation area.

[0027] For example, if the content is a still image and the prompt is estimated to be one that does not perform "generation that is significantly contrary to the content (facts)," such as "remove noise" or "adjust brightness," the prompt analysis unit 114 estimates that "nothing has been generated (there is no generation area)."

[0028] Furthermore, if the prompt analysis unit 114 estimates that a prompt performs "generation that is significantly contrary to the content (facts)," such as adding or erasing objects, it obtains "large" (or a value corresponding to "large") as the degree of generation of the generation area corresponding to that prompt. If the prompt estimates that a prompt performs "generation that is not significantly contrary to the content (facts)," such as changing the color of objects, it obtains "small" (or a value corresponding to "small") as the degree of generation of the generation area corresponding to that prompt. In this way, the prompt analysis unit 114 obtains a degree of generation corresponding to the type of processing (the type of processing performed by the generation model) corresponding to the prompt.

[0029] Furthermore, the method for determining the degree of generation from the type of processing corresponding to the prompt (the type of content generation performed by the generation model) is not limited to a specific method. For example, the prompt analysis unit 114 may use a table that associates the type of processing corresponding to the prompt (the type of content generation performed by the generation model) with the degree of generation to obtain the degree of generation corresponding to the type of processing corresponding to the prompt (the type of content generation performed by the generation model).

[0030] The model analysis unit 115 obtains "information about the generation model" contained in the information acquired in step S202. For example, if the "information about the generation model" is name information representing the name of the generation model, the model analysis unit 115 obtains "feature information representing the characteristics of the generation model" associated with the name information from a database held by the information processing device 100 or a database provided outside the information processing device 100. Also, for example, if the "information about the generation model" is feature information, the model analysis unit 115 obtains that feature information. Based on this feature information, the model analysis unit 115 estimates that there is a high probability that the area in the content generated by the generation model is a generated area.

[0031] Feature information includes, for example, information that describes the tendency of the objects to be generated or the range of the area to be generated. For example, if the feature information indicates that "the generation model for still images has a feature that extends the image range smoothly outside the original image (Outpainting function)," the model analysis unit 115 can estimate that the area outside the still image is more likely to be a generated area. Also, if the feature information indicates that "the generation model for images has a feature that removes noise and adjusts brightness appropriately," the model analysis unit 115 estimates that "nothing has been generated."

[0032] In this way, the analysis unit 112 can identify / estimate the generation area in the content and obtain the degree of generation of the generation area. Note that the method for identifying / estimated the generation area in the content is not limited to the method described above, nor is the number of methods limited to the method described above. In other words, the configuration of the analysis unit 112 shown in Figure 2 is just one example and is not limited to this configuration.

[0033] In step S205, the analysis unit 112 determines whether the generation region has been identified or estimated as a generation region based on the results of the analysis in step S204. If, as a result of this determination, the generation region has been identified or estimated as a generation region, the process proceeds to step S206. On the other hand, if the generation region has not been identified or estimated as a generation region, the process according to the flowchart in Figure 3 ends.

[0034] In step S206, the region detection unit 116 detects the identified generation region and the region estimated to be a generation region as detection regions, based on the results of the analysis in step S204.

[0035] For example, if the content is a still image and the region analysis unit 113 identifies a generated region in the still image through the above analysis, the region detection unit 116 detects the generated region as a detection region.

[0036] Furthermore, for example, if the content is a still image, and the prompt analysis unit 114 estimates that the area in the lower right position of the still image is likely to be the generation area based on the prompt "Draw a flower in the lower right position of the still image", the area detection unit 116 will detect the "flower" area in the lower right position of the still image as a detection area using an object detection method such as YOLO (You Only Look Once). If there are other "flower" areas in the still image other than the lower right position, the area detection unit 116 will also detect those areas as detection areas.

[0037] For example, if the content is a still image, and the prompt analysis unit 114 estimates that the area to the left of the group of objects contained in the still image is likely to be the generation area based on the prompt "Remove the person on the left in the still image", the area detection unit 116 detects an object that appears to be the main subject from the still image (for example, the object closest to the center of the still image, or the object with the largest size). The area detection unit 116 then detects a predetermined area to the left of the detected object as the detection area.

[0038] Another example is when the model analysis unit 115 recognizes that the generative model used has the characteristic of seamlessly extending the image range outside the original image; in this case, the region detection unit 116 detects the region outside the image as the detection region.

[0039] In step S207, the region presentation unit 117 displays the detected region detected by the region detection unit 116 in step S206 on the display screen. The display screen may be a display screen of the information processing device 100, or a display screen of an external device that can communicate with the information processing device 100.

[0040] The following describes an example of presenting a detection region. The following describes a case in which the still image 310 shown in Figure 4(b) is generated as content from the still image 300 shown in Figure 4(a) by the generation model. Still image 310 is a still image generated by the generation model from still image 300 based on the first prompt "remove the person on the left in the still image" and the second prompt "draw a flower in the lower right of the still image". In other words, the generation model removed the subject 301 from still image 300 based on the first prompt and generated a new still image as still image 310 by placing a flower 311 in the lower right of still image 300 based on the second prompt. Figure 5 shows an example of the display of the presentation window 400 as a GUI displayed on the display screen as a result of processing according to the flowchart in Figure 3 when the content is such a still image 310. The display control of the presentation window 400 is performed by the region presentation unit 117. The still image 310, which is the content generated by the generation model, is displayed in region 410.

[0041] The detected region 412 is a generated region generated by the generation model in response to the prompt "Remove the person on the left in the still image 300," and is detected as a detected region by the region detection unit 116. The detected region 412 is a region with multiple generation possibilities, based on the judgment that a person object that was somewhere in the large space to the left of the main subject has been removed. In Figure 5, the higher the generation possibility, the darker the color displayed, and the detected region 412 is displayed in a different color for each generation possibility.

[0042] The more specifically a prompt indicates the generation region, the higher the probability of generation corresponding to that prompt. For example, if the prompt includes an object name, the probability of generation corresponding to that prompt is high. The prompt analysis unit 114 can calculate the probability of generation according to the words and number of words included in the prompt. Note that the method for calculating the probability of generation is not limited to a specific calculation method.

[0043] The detected region 411 is a generated region generated by the generation model in response to the prompt "An object named 'flower' has been generated in the lower right position of the still image 310," and is a region detected as a detected region by the region detection unit 116. Since this prompt includes the object name "flower," the probability of generation corresponding to this prompt is high, and therefore the detected region 411 is presented in a darker color.

[0044] In this way, by presenting areas that are more likely to have been generated in a more prominent color, the likelihood of generation in those areas can be intuitively grasped. Therefore, the colors corresponding to generation probability are not limited to specific colors, nor are the display methods limited to specific methods, as long as this objective is achieved.

[0045] Area 420 is provided with adjustment bars 421 and 422. The user can move adjustment bar 421 left or right by operating an operation unit (not shown). Area 410 displays the detection area of ​​generation probability that belongs to the range of generation probability corresponding to the position of adjustment bar 421, starting from the lowest generation probability (lowest generation probability) at the lower end (left end) of the range that adjustment bar 421 can move. For example, if you want to display only areas that have definitely been generated in area 410, you can move adjustment bar 421 to near the right end (near "high") so that areas with a low probability of generation are not displayed in area 410.

[0046] The user can move the adjustment bar 422 left or right by operating an unillustrated control unit. The detection area for generation degrees that belong to the range of generation degrees corresponding to the position of the adjustment bar 422, from the generation degree at the lower end (left end) of the range that the adjustment bar 422 can move (the smallest generation degree), is displayed in area 410. If you want to display only areas with a large generation degree in area 410, you can move the adjustment bar 422 to near the right end (near "large"), and areas with a small generation degree will no longer be displayed.

[0047] In region 430, the original image estimated backward from the still image 310 based on the results of the analysis in step S204 is displayed. The detected region 411 is estimated to be a region generated according to the prompt "Draw a flower in the lower right." Therefore, the region presentation unit 117 predicts that there were no objects in the region corresponding to the detected region 411 in the original image, and uses the image generation model to generate an image in which the detected region 411 of the still image 310 is filled in as the first image. The detected region 412 is also estimated to be a region generated according to the prompt "Erase the person on the left." Therefore, the region presentation unit 117 predicts that a person object existed in the region corresponding to the detected region 412 in the original image, and uses the image generation model to generate a second image in which a person image is generated in the region corresponding to the detected region 412 in the first image, and uses this image generation model to generate a second image as the original image. The region presentation unit 117 then displays the original image (second image) generated in this way in region 430.

[0048] By performing this process, even if the original image cannot be restored, a predicted source image with a meaning similar to the original image can be generated. On the other hand, if the acquisition unit 111 can only acquire the position information of the detection area instructed when the content was generated, it is possible to estimate the detection area within the still image 310, but there is not enough information to create a predicted source image. Therefore, if it is difficult to create a predicted source image, it is not necessary to display the predicted source image in area 430.

[0049] When viewing only the still image 310, it is not immediately clear whether or not it is an image generated by the generative model, and even if it is recognized as an image generated by the generative model, it is difficult to determine which region has been generated. However, according to this embodiment, even if it is not immediately clear whether the content is generated, or if the generated area is unknown, the generated region or a region that is likely to be a generated region is detected and presented, making it possible to intuitively grasp the generated region.

[0050] In the first embodiment, the detection area was displayed in a color corresponding to the possibility of generation. However, this is merely one example of displaying the detection area in a display format corresponding to the possibility of generation. For example, the thickness and shape of the frame of the detection area may be changed according to the possibility of generation.

[0051] <Example 1> The recording unit 109 may associate the content acquired in step S201 with the information regarding the detected area presented in area 410 and store it in the non-volatile memory of the information processing device 100 or in the memory of an external server device that can communicate with the information processing device 100. This makes it possible to check the generated area within the content again without having to re-execute the process according to the flowchart in Figure 3.

[0052] <Modification 2> In step S206, the cropping unit 118 crops out the detected region detected by the region detection unit 116. If the content is a still image, the cropping unit 118 crops out a partial image within the region detected by the region detection unit 116, and the region presentation unit 117 presents the partial image and the image of the region of the still image other than the partial image separately.

[0053] The cropping unit 118 may also be configured to crop a portion of the image outside the region detected by the region detection unit 116, and even in that case, the operation of the region presentation unit 117 remains the same.

[0054] This allows for an intuitive understanding of the generated area, even when it is unclear at first glance whether the content is generated or where it was generated. The system detects and presents the generated area or an area that is likely to be generated.

[0055] Furthermore, the recording unit 109 may store the detected area and / or non-detected area in the non-volatile memory of the information processing device 100 or in the memory of an external server device that can communicate with the information processing device 100. In this way, if you want to check the generated area in the content again, you can check the generated area or non-generated area without having to re-execute the process according to the flowchart in Figure 3.

[0056] [Second Embodiment] Each functional unit of the information processing device 100 in Figure 2 may be implemented in hardware or in software (computer program). In the latter case, a computer device capable of executing such a computer program can be applied to the information processing device 100. An example of the hardware configuration of such a computer device will be explained using the block diagram in Figure 1.

[0057] The CPU 101 executes various processes using computer programs and data stored in the RAM 102. In doing so, the CPU 101 controls the operation of the entire computer system and executes or controls the various processes described as being performed by the information processing device 100.

[0058] RAM 102 has an area for storing computer programs and data loaded from ROM 103 and storage device 104, and a work area used by the CPU 101 when executing various processes. In this way, RAM 102 can provide various areas as appropriate.

[0059] ROM103 stores configuration data for the computer device, computer programs and data related to the startup of the computer device, computer programs and data related to the basic operation of the computer device, and so on.

[0060] The storage device 104 is an example of a large-capacity non-volatile memory such as a hard disk. The storage device 104 stores the OS (operating system), computer programs and data for executing or controlling the CPU 101 for various processes described as being performed by the information processing device 100, and so on. For example, the recording unit 119 can store various types of information in the storage device 104. The recording medium 104 may also be a device that reads and writes information to a memory card such as an SD card that can be attached and removed.

[0061] The operation unit 105 is a user interface such as a keyboard, mouse, or touch panel, which allows the user to input various instructions and information to the computer device through its operation. For example, the user can operate the operation unit 105 to operate the presentation window 400 in Figure 5 or to input prompts to the computer device.

[0062] The display unit 106 has an LCD screen or a touch panel screen and can display the processing results of the CPU 101 as images, text, etc. For example, the display unit 106 can display the presentation window 400 shown in Figure 5. The display unit 106 may also be a projection device such as a projector that projects images, text, etc.

[0063] The CPU 101, RAM 102, ROM 103, storage device 104, operation unit 105, and display unit 106 are all connected to the system bus 130. Note that the configuration shown in Figure 1 is an example of a hardware configuration for a computer device applicable to the information processing device 100, and is not limited to this configuration. For example, a network interface may be further connected to the system bus 130, in which case the computer device can communicate data with an external server device via the network interface.

[0064] The numerical values, processing timing, processing order, processing entity, data (information) structure / acquisition method / destination / source / storage location, etc., used in the above embodiment are given as examples for the purpose of providing a concrete explanation, and are not intended to limit the scope to such examples.

[0065] Furthermore, some or all of the embodiments described above may be used in appropriate combinations. Alternatively, some or all of the embodiments described above may be used selectively.

[0066] (Other embodiments) The present invention can also be realized by supplying a program that implements one or more of the functions of the above-described 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.

[0067] The inventions described herein include the following information processing devices, information processing methods, and computer programs. (Item 1) A detection means for detecting, based on information related to content generation, regions generated by a generative model and / or regions presumed to have been generated by the generative model in content generated by the generative model, A presentation means that presents the region detected by the detection means. An information processing device characterized by comprising: (Item 2) The information processing apparatus according to item 1, characterized in that the detection means detects a region in the content corresponding to the location information if the information includes location information of a generated region generated by the generation model. (Item 3) The information processing apparatus according to item 1 or 2, characterized in that the detection means detects a region in the content that is estimated based on the wording contained in the prompt, if the information includes a prompt designated for the generation of the content. (Item 4) The information processing apparatus according to any one of items 1 to 3, characterized in that the detection means detects a region in the content corresponding to the information about the generation model when the information includes information about the generation model. (Item 5) The information processing device according to any one of items 1 to 4, characterized in that the display means displays the region detected by the detection means in a display format corresponding to the possibility that it is a region generated by the generation model. (Item 6) The information processing device according to any one of items 1 to 5, characterized in that the display means displays a region from among the regions detected by the detection means according to a specified possibility. (Item 7) The information processing device according to any one of items 1 to 6, characterized in that the display means displays a region according to the type of processing performed by the generation model. (Item 8) The information processing apparatus according to item 7, characterized in that the type of processing performed by the generation model includes whether or not it is the generation of an area that is significantly contrary to the content. (Item 9) moreover, An information processing apparatus according to any one of items 1 to 8, characterized by comprising means for associating the content with information relating to the region detected by the detection means and storing it in memory. (Item 10) An information processing device according to any one of items 1 to 9, characterized by presenting a region detected by the detection means and a region in the content that is different from said region. (Item 11) moreover, The information processing apparatus according to any one of items 1 to 10, characterized by comprising means for storing in memory an area and / or an area in the content different from the area detected by the detection means. (Item 12) The information processing device described in any one of items 1 to 11, characterized in that the information related to the generation of the content is metadata for the content. (Item 13) An information processing method performed by an information processing device, The detection means of the information processing device includes a detection step of detecting, in content generated by the generation model, regions generated by the generation model and / or regions presumed to have been generated by the generation model, based on information related to content generation, The presentation means of the information processing device performs a presentation step of presenting the region detected in the detection step. An information processing method characterized by comprising: (Item 14) A computer program that causes a computer to function as one of the means of an information processing device described in any one of items 1 through 12.

[0068] The invention is not limited to the embodiments described above, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, claims are attached to disclose the scope of the invention. [Explanation of Symbols]

[0069] 100: Information processing device 110: Acquisition unit 111: Acquisition unit 112: Analysis unit 113: Region analysis unit 114: Prompt analysis unit 115: Model analysis unit 116: Region detection unit 117: Region presentation unit 118: Extraction unit 119: Recording unit

Claims

1. A detection means for detecting, based on information related to content generation, regions generated by a generative model and / or regions presumed to have been generated by the generative model in content generated by the generative model, A presentation means that presents the region detected by the detection means. An information processing device characterized by comprising:

2. The information processing apparatus according to claim 1, wherein the detection means detects a region in the content corresponding to the location information if the information includes location information of a generated region generated by the generation model.

3. The information processing apparatus according to claim 1, wherein the detection means detects a region in the content that is estimated based on the wording contained in the prompt, if the information includes a prompt designated for the generation of the content.

4. The information processing apparatus according to claim 1, wherein the detection means detects a region in the content corresponding to the information about the generation model when the information includes information about the generation model.

5. The information processing apparatus according to claim 1, characterized in that the display means displays the region detected by the detection means in a display format corresponding to the possibility that it is a region generated by the generation model.

6. The information processing apparatus according to claim 1, characterized in that the display means displays a region from among the regions detected by the detection means according to a specified possibility.

7. The information processing apparatus according to claim 1, characterized in that the display means displays a region according to the type of processing performed by the generation model.

8. The information processing apparatus according to claim 7, characterized in that the type of processing performed by the generation model includes whether or not it is the generation of an area that is significantly contrary to the content.

9. moreover, The information processing apparatus according to claim 1, further comprising means for associating the content with information relating to the region detected by the detection means and storing it in memory.

10. The information processing apparatus according to claim 1, characterized in that it presents a region detected by the detection means and a region in the content that is different from the region detected.

11. moreover, The information processing apparatus according to claim 1, further comprising means for storing in memory an area and / or an area in the content different from the area detected by the detection means.

12. The information processing apparatus according to claim 1, characterized in that the information related to the generation of the content is metadata for the content.

13. An information processing method performed by an information processing device, The detection means of the information processing device includes a detection step of detecting, in content generated by the generation model, regions generated by the generation model and / or regions presumed to have been generated by the generation model, based on information related to content generation, The presentation means of the information processing device performs a presentation step of presenting the region detected in the detection step. An information processing method characterized by comprising:

14. A computer program for causing a computer to function as one of the means of an information processing apparatus described in any one of claims 1 to 12.

Images