Automatic photographing screening method and device and electronic equipment

Abstract

The invention provides an automatic photographing screening method and device, electronic equipment and a storage medium, and relates to the technical field of image processing, and the method comprises the steps: carrying out the analysis of a collected image, and obtaining an analysis result; under the condition that the analysis result meets a preset triggering condition, at least two photos are shot; and screening the at least two photos to obtain a target photo, so that a user can obtain a satisfactory photo, thereby obtaining a wonderful moment.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to an automatic photo selection method, apparatus, and electronic device. Background Technology

[0002] Cameras are widely used in various smart products, and most of them are used to take pictures by manual control. However, many beautiful moments are often fleeting and cannot be captured manually. Moreover, due to factors such as lighting, latency, and hand shakiness, the photos taken do not meet the user's requirements.

[0003] Therefore, how to enable users to obtain satisfactory photos and capture wonderful moments is an urgent problem to be solved. Summary of the Invention

[0004] This invention provides an automatic photo selection method, apparatus, and electronic device to solve the problem of how to enable users to obtain satisfactory photos and capture wonderful moments.

[0005] This invention provides an automatic photo filtering method, comprising: The acquired images are analyzed to obtain the analysis results; If the analysis results meet the preset triggering conditions, at least two photos will be taken; The target photo is obtained by filtering the at least two photos.

[0006] According to an automatic image screening method provided by the present invention, the step of analyzing the acquired images to obtain analysis results includes: The acquired image is analyzed from at least one first dimension to obtain the analysis result; the first dimension includes any one of the following: scene, object, and lighting.

[0007] According to an automatic image screening method provided by the present invention, the step of analyzing the acquired image from at least one first dimension to obtain the analysis result includes: An image analysis model is used to analyze the acquired image from at least one first dimension to obtain the analysis result.

[0008] According to an automatic photo selection method provided by the present invention, the triggering conditions include at least one of the following: target scene, object features, light and shadow features, and composition logic.

[0009] According to an automatic photo selection method provided by the present invention, taking at least two photos includes: Based on a preset shooting cycle, take at least two photos; or, Based on preset rules, at least two photos are taken.

[0010] According to an automatic photo filtering method provided by the present invention, the step of filtering the at least two photos to obtain a target photo includes: Based on a preset screening period, a large model is used to screen the at least two photos to obtain the target photo.

[0011] According to an automatic photo filtering method provided by the present invention, the step of filtering at least two photos using a large model to obtain the target photo includes: For each of the at least two photos, a large model is used to analyze and score from at least one second dimension to obtain a score for each second dimension; The target score is determined based on the scores of each of the second dimensions; The photos corresponding to the top N target scores from each of the target scores are determined as the target photos.

[0012] According to an automatic photo screening method provided by the present invention, the second dimension includes any one of the following: aesthetic evaluation, content value, technical quality, facial expressions, actions, emotions, and whether the scene is the same.

[0013] The present invention also provides an automatic photo screening device, comprising: The analysis module is used to analyze the acquired images and obtain the analysis results; The camera module is used to take at least two photos when the analysis results meet preset triggering conditions. The filtering module is used to filter the at least two photos to obtain the target photo.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the automatic photo screening method as described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the automatic photo screening method as described above.

[0016] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the automatic photo screening method as described above.

[0017] The automatic photo selection method, device, electronic device, and storage medium provided by this invention analyze the acquired images to obtain analysis results; when the analysis results meet preset triggering conditions, at least two photos are taken; the at least two photos are selected to obtain target photos, enabling users to obtain satisfactory photos and capture wonderful moments. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is one of the flowcharts of the automatic photo filtering method provided by the present invention.

[0020] Figure 2 This is the second flowchart of the automatic photo filtering method provided by the present invention.

[0021] Figure 3 This is the third flowchart of the automatic photo filtering method provided by the present invention.

[0022] Figure 4 This is a schematic diagram of the automatic photo screening device provided by the present invention.

[0023] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] The following is combined Figures 1-3 The automatic photo filtering method of the present invention is described.

[0026] Figure 1 This is one of the flowcharts illustrating the automatic photo filtering method provided by the present invention, such as... Figure 1 As shown, the method includes the following steps 101-103.

[0027] Step 101: Analyze the acquired images and obtain the analysis results.

[0028] It should be noted that the photo filtering method provided by the present invention can be applied to automatic photo shooting scenarios. The subject executing the method can be a photo filtering device, such as an electronic device, or a control module in the photo filtering device for executing the photo filtering method.

[0029] Specifically, images can be acquired in real time using a camera, and then analyzed in real time to obtain analysis results. The analysis results include at least one of the following: the scene corresponding to the image, the objects in the image, and the lighting and shadow features in the image. The scene corresponding to the image includes at least one of the following: indoor scene, outdoor scene, sports scene, and music scene; other scenes are also acceptable without any limitation. The objects in the image include at least one of the following: people, a frontal portrait of a person, a profile portrait of a person, animals, a coexistence of people and animals, and specific objects; other objects are also acceptable without any limitation. The lighting and shadow features in the image include at least one of the following: sunset, sunrise, rainbow, aurora, plant lighting and shadow, and architectural lighting and shadow; other lighting and shadow features are also acceptable without any limitation.

[0030] Step 102: If the analysis results meet the preset triggering conditions, take at least two photos.

[0031] Specifically, trigger conditions are preset so that the camera can be automatically triggered to take at least two photos when the analysis results meet the preset trigger conditions.

[0032] Optionally, the triggering conditions include at least one of the following: target scene, object features, lighting features, and composition logic. That is, when the analysis result satisfies at least one of the following, the camera can be automatically triggered to take at least two photos. The target scene can be an indoor scene, an outdoor scene, a sports scene, a music scene, or other scenes. Object features include at least one of the following: a person, a frontal portrait of a person, a side profile of a person, an animal, a person and an animal coexisting, a specific object, or other objects. Lighting features include at least one of the following: sunset, sunrise, rainbow, aurora, plant lighting, and architectural lighting, or other lighting features. Composition logic includes at least one of the following: rule of thirds, golden ratio, symmetrical composition, diagonal composition, frame composition, central composition, leading lines composition, foreground composition, triangular composition, and use of negative space.

[0033] Step 103: Filter the at least two photos to obtain the target photo.

[0034] Specifically, after taking at least two photos, the user can filter the photos to obtain the target photo that meets their requirements.

[0035] The automatic photo selection method provided by this invention analyzes the collected images to obtain analysis results; when the analysis results meet preset triggering conditions, at least two photos are taken; the at least two photos are selected to obtain target photos, enabling users to obtain satisfactory photos and capture wonderful moments.

[0036] Optionally, the specific implementation of step 101 above includes: The acquired image is analyzed from at least one first dimension to obtain the analysis result; the first dimension includes any one of the following: scene, object, and lighting.

[0037] Specifically, the first dimension includes any one of the following: scene, object, and lighting. A scene includes at least one of the following: indoor scene, outdoor scene, motion scene, and music scene, but may also be other scenes without limitation. An object includes at least one of the following: person, frontal portrait of a person, profile portrait of a person, animal, coexistence of person and animal, and specific object, but may also be other objects without limitation. Analysis results can be obtained by analyzing the acquired images from at least one of the first dimensions.

[0038] Optionally, the analysis of the acquired image from at least one first dimension to obtain the analysis result includes: An image analysis model is used to analyze the acquired image from at least one first dimension to obtain the analysis result.

[0039] Specifically, image analysis models can be pre-trained models. From a scene perspective, these models can be traditional bag-of-words models, image scene recognition models, image scene classification models, and image semantic analysis models. Among these, image scene recognition models, image scene classification models, and image semantic analysis models can be convolutional neural networks (CNNs), recurrent neural networks (RNNs), convolutional recurrent neural networks, or CNN-based image classification models, such as VGG, ResNet, and Inception. From an object perspective, image analysis models can be object detection models and face recognition models, such as region-based convolutional neural networks (RCNNs), YOLO, and single-shot multibox detectors (SSDs). From a lighting perspective, image analysis models can be Lambert lighting models, Phong lighting models, Blinn-Phong models, three-color decay models, diffusion models, and attention mechanism models.

[0040] The automatic photo filtering method provided by this invention uses an image analysis model to perform real-time analysis on images from at least one of the first dimensions: scene, object, and light and shadow. This accurately obtains the analysis results, and then determines whether preset triggering conditions are met based on the analysis results, triggering the shooting of photos. This achieves the filtering of target photos, enabling users to obtain satisfactory photos and capture wonderful moments, thereby improving the efficiency of target photo filtering and enhancing the user experience.

[0041] Optionally, the specific implementation of step 102 above includes: Based on a preset shooting cycle, at least two photos are taken; or, based on preset rules, at least two photos are taken.

[0042] Specifically, a shooting cycle is preset, for example, shooting once every 5 seconds, once every 2 seconds, or once every 30 seconds. If a user wants to take 100 photos, they can shoot once every 5 seconds, once every 2 seconds, or once every 30 seconds until all 100 photos are taken. A preset rule could be to take N photos continuously at once, then wait M seconds before taking another N photos continuously, where N is greater than 1 and M is greater than 0. For example, a preset rule could be to take 5 photos continuously at once, then wait 10 seconds before taking another 5 photos continuously. If a user wants to take 100 photos, they can first take 5 photos continuously at once, then wait 10 seconds before taking another 5 photos continuously, then wait another 10 seconds before taking another 5 photos continuously, and so on, until all 100 photos are taken.

[0043] The photo selection method provided by this invention achieves photo shooting based on a preset shooting cycle or preset rules, without human intervention, and can achieve the target number of photos required by the user, thereby enabling the selection of target photos and allowing the user to obtain satisfactory photos and capture wonderful moments.

[0044] Optionally, the step of filtering the at least two photos to obtain the target photo includes: Based on a preset screening period, a large model is used to screen the at least two photos to obtain the target photo.

[0045] Specifically, at least two photos taken within a preset screening period are selected. For example, if the preset screening period is one day and 158 photos are taken in one day, the 158 photos are analyzed, sorted, and selected from the second dimension.

[0046] Input at least two photos and a prompt word into a large model. The large model can then filter the photos based on the prompt word to obtain the target photo. For example, the large model could be Deepseek.

[0047] Optionally, the step of using a large model to filter the at least two photos to obtain the target photo includes: For each of the at least two photos, a large model is used to analyze and score from at least one second dimension to obtain a score for each second dimension; based on the scores for each second dimension, a target score is determined; the photos corresponding to the top N target scores are determined as the target photos.

[0048] Optionally, the second dimension includes any of the following: aesthetic evaluation, content value, technical quality, character expressions, character actions, character emotions, and whether the scene is the same.

[0049] Specifically, aesthetic evaluation assesses the aesthetic quality of each photograph from dimensions such as composition, lighting, and color. Content value refers to the content containing the target object, such as people, animals, or specific targets, where the target object is a specific object, such as a car and / or a building. Technical quality refers to photographs with technical defects such as blurriness, out-of-focus, or overexposure. Facial expression is evaluated by detecting facial key points and calculating the eye aspect ratio (EAR) value to assess the vividness of facial expressions. Facial movement is evaluated using optical flow algorithms to assess the amplitude, speed, and smoothness of movements, selecting dynamic and exciting moments. Facial emotion is analyzed by combining at least one of facial expressions, body language, and audio content to determine the emotional state.

[0050] For each of at least two photos, a large model is used to analyze and score it from at least one second dimension, obtaining a score for each second dimension. Then, based on the scores for each second dimension, the scores for each second dimension are added together to obtain the target score. Alternatively, the score weights for each second dimension can be pre-set, and the score for each second dimension is multiplied by its corresponding weight, resulting in a product. The target scores are then sorted in descending order, and the top N target scores are determined. The photos corresponding to the top N target scores are then identified as the target photos.

[0051] It should be noted that when the content of a certain second dimension triggers the preset sensitive content, the overall rejection logic will be triggered, and the photo will not participate in the scoring and screening.

[0052] Alternatively, photos with the same scene can be grouped together, then the target score for each photo in each group can be determined, and then the target scores can be sorted in descending order. The top N target scores can be determined from the sorted target scores, and the photos corresponding to the top N target scores can be determined as target photos.

[0053] The photo filtering method provided by this invention analyzes and scores from at least one second dimension using a large model to obtain scores for each second dimension; based on the scores for each second dimension, a target score is determined; the photos corresponding to the top N target scores are identified as target photos, thereby achieving the filtering of target photos. This allows users to obtain satisfactory photos, capture wonderful moments, improve the efficiency of target photo filtering, and thus enhance the user experience.

[0054] Figure 2 This is the second flowchart of the automatic photo filtering method provided by the present invention, as shown below. Figure 2 As shown, the method includes steps 201-205.

[0055] Step 201: Acquire images.

[0056] Step 202: Analyze the acquired images to obtain analysis results. The analysis results include at least one of the following: the scene corresponding to the image, the objects in the image, and the lighting and shadow features in the image.

[0057] Step 203: Determine whether the analysis result meets the preset trigger conditions. The trigger conditions include at least one of the following: target scene, object features, lighting and shadow features, and composition logic. If the analysis result meets the preset trigger conditions, proceed to step 204; if the analysis result does not meet the preset trigger conditions, proceed to step 202.

[0058] Step 204: Take at least two photos.

[0059] Step 205: Filter at least two photos to obtain the target photo.

[0060] The automatic photo filtering method provided by this invention analyzes the collected images in real time to obtain analysis results; determines whether the analysis results meet preset trigger conditions; if the preset trigger conditions are met, takes at least two photos; and then filters the at least two photos to obtain target photos, enabling users to obtain satisfactory photos, thereby capturing wonderful moments, improving the efficiency of target photo filtering, and thus enhancing the user experience.

[0061] Figure 3 This is the third flowchart of the automatic photo filtering method provided by the present invention, as shown below. Figure 3 As shown, the method includes steps 301-307.

[0062] Step 301: Acquire images.

[0063] Step 302: Using an image analysis model, analyze the acquired image from at least one first dimension to obtain the analysis result. The image analysis model can be a pre-trained model. From the scene dimension, the image analysis model can be a traditional bag-of-words model, an image scene recognition model, an image scene classification model, or an image semantic analysis model. The image scene recognition model, image scene classification model, and image semantic analysis model can be a convolutional neural network (CNN), a recurrent neural network (RNN), a convolutional recurrent neural network, or an image classification model based on CNN, such as VGG, ResNet, Inception, etc. From the object dimension, the image analysis model can be an object detection model or a face recognition model, such as a region-based convolutional neural network (RCNN), YOLO, or a single-shot multibox detector (SSD). From the lighting dimension, the image analysis model can be a Lambert lighting model, a Phong lighting model, a Blinn-Phong model, a three-color decay model, a diffusion model, or an attention mechanism model. The first dimension includes any one of the following: scene, object, and lighting. A scene includes at least one of the following: indoor scene, outdoor scene, motion scene, and music scene, but may also be other scenes without limitation. An object includes at least one of the following: person, a frontal portrait of a person, a profile portrait of a person, an animal, a person and an animal coexisting, and a specific object, but may also be other objects without limitation.

[0064] Step 303: Determine whether the analysis result meets the preset triggering conditions. The triggering conditions include at least one of the following: target scene, object features, lighting features, and composition logic. That is, when the analysis result meets at least one of the following, the camera can be automatically triggered to take at least two photos. The target scene can be an indoor scene, an outdoor scene, a sports scene, a music scene, or other scenes. Object features include at least one of the following: people, a frontal portrait of a person, a side profile of a person, animals, people and animals coexisting, a specific object, or other objects. Lighting features include at least one of the following: sunset, sunrise, rainbow, aurora, plant lighting, and architectural lighting, or other lighting features. Composition logic includes at least one of the following: rule of thirds, golden ratio, symmetrical composition, diagonal composition, frame composition, central composition, leading lines composition, foreground composition, triangular composition, and use of negative space. If the analysis results meet the preset trigger conditions, proceed to step 304; if the analysis results do not meet the preset trigger conditions, proceed to step 302.

[0065] Step 304: Take at least two photos based on a preset shooting cycle or preset rules. Specifically, a shooting cycle is preset, for example, taking a photo every 5 seconds, every 2 seconds, or every 30 seconds. If the user wants to take 100 photos, they can take a photo every 5 seconds, every 2 seconds, or every 30 seconds until all 100 photos are taken. A preset rule can be to take N photos continuously at once, then take another N photos after an M-second interval, where N is greater than 1 and M is greater than 0. For example, a preset rule could be to take 5 photos continuously at once, then take another 5 photos after a 10-second interval, then take another 5 photos after a 10-second interval, and so on, until all 100 photos are taken.

[0066] Step 305: For each of the at least two photos, based on a preset screening period, a large model is used to analyze and score from at least one second dimension to obtain a score for each second dimension. Specifically, the second dimension includes any one of the following: aesthetic evaluation, content value, technical quality, facial expression, facial movement, facial emotion, and whether the scene is the same. Aesthetic evaluation refers to assessing the aesthetic quality of each photo from dimensions such as composition, lighting, and color. Content value refers to content containing a target object, such as content containing people, animals, or specific targets, where a specific target is a specific object, such as a car and / or a building. Technical quality refers to photos with technical defects such as blurriness, out-of-focus, or overexposure. Facial expression refers to assessing the vividness of facial expressions through facial key point detection and eye aspect ratio (EAR) value calculation. Facial movement refers to using an optical flow algorithm to assess the amplitude, speed, and smoothness of movements, filtering for dynamic and exciting moments. Facial emotion refers to analyzing the emotional state by combining at least one of facial expression and body language. For each of the at least two photos, a large model is used to analyze and score from at least one second dimension to obtain a score for each second dimension.

[0067] Step 306: Determine the target score based on the scores of each of the second dimensions. Specifically, based on the scores of each of the second dimensions, add the scores of each second dimension together to obtain the target score. Alternatively, the score weights of each second dimension can be preset, the score of each second dimension can be multiplied by its corresponding weight, and the products of each second dimension can be added together to obtain the target score.

[0068] Step 307: Determine the photos corresponding to the top N target scores from each of the target scores, and designate them as the target photos. Specifically, sort the target scores in descending order, determine the top N target scores from the sorted target scores, and designate the photos corresponding to the top N target scores as the target photos.

[0069] The automatic photo filtering method provided by this invention analyzes the acquired images from at least one first dimension using an image analysis model to obtain analysis results; determines whether the analysis results meet preset trigger conditions; if the preset trigger conditions are met, takes at least two photos based on a preset shooting cycle or preset rules; then combines a large model to analyze and score from at least one second dimension to obtain scores for each second dimension; determines target scores based on the scores for each second dimension; and identifies the photos corresponding to the top N target scores as target photos, thereby achieving target photo filtering. This allows users to obtain satisfactory photos, capture wonderful moments, improve the efficiency of target photo filtering, and enhance the user experience.

[0070] The automatic photo filtering device provided by the present invention is described below. The automatic photo filtering device described below can be referred to in correspondence with the automatic photo filtering method described above.

[0071] Figure 4 This is a schematic diagram of the automatic photo-screening device provided by the present invention, as shown below. Figure 4 As shown, the photo screening device 400 includes: an analysis module 401, a shooting module 402, and a screening module 403; wherein, Analysis module 401 is used to analyze the acquired images and obtain analysis results; The shooting module 402 is used to take at least two photos when the analysis results meet the preset triggering conditions; The filtering module 403 is used to filter the at least two photos to obtain the target photo.

[0072] The automatic photo filtering device provided by this invention analyzes the collected images to obtain analysis results; when the analysis results meet preset triggering conditions, it takes at least two photos; and filters the at least two photos to obtain target photos, enabling users to obtain satisfactory photos and capture wonderful moments.

[0073] Optionally, the analysis module 401 is specifically used for: The acquired image is analyzed from at least one first dimension to obtain the analysis result; the first dimension includes any one of the following: scene, object, and lighting.

[0074] Optionally, the analysis module 401 is further configured to: An image analysis model is used to analyze the acquired image from at least one first dimension to obtain the analysis result.

[0075] Optionally, the triggering condition includes at least one of the following: target scene, object features, lighting features, and composition logic.

[0076] Optionally, the shooting module 402 is specifically used for: Based on a preset shooting cycle, take at least two photos; or, Based on preset rules, at least two photos are taken.

[0077] Optionally, the filtering module 403 is specifically used for: Based on a preset screening period, a large model is used to screen the at least two photos to obtain the target photo.

[0078] Optionally, the filtering module 403 is further configured to: For each of the at least two photos, a large model is used to analyze and score from at least one second dimension to obtain a score for each second dimension; The target score is determined based on the scores of each of the second dimensions; The photos corresponding to the top N target scores from each of the target scores are determined as the target photos.

[0079] Optionally, the second dimension includes any of the following: aesthetic evaluation, content value, technical quality, character expressions, character actions, character emotions, and whether the scene is the same.

[0080] Figure 5 This is a schematic diagram of the physical structure of an electronic device provided by the present invention, such as... Figure 5 As shown, the electronic device 500 may include a processor 510, a communications interface 520, a memory 530, and a communication bus 540, wherein the processor 510, communications interface 520, and memory 530 communicate with each other via the communication bus 540. The processor 510 can call logical instructions in the memory 530 to execute an automatic image selection method, which includes: analyzing the acquired images to obtain analysis results; taking at least two photos when the analysis results meet preset trigger conditions; and selecting the at least two photos to obtain target photos.

[0081] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0082] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the automatic photo selection method provided by the above methods. The method includes: analyzing the acquired images to obtain analysis results; taking at least two photos when the analysis results meet preset triggering conditions; and selecting the at least two photos to obtain target photos.

[0083] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the automatic photo selection method provided by the above methods. The method includes: analyzing the acquired images to obtain analysis results; taking at least two photos when the analysis results meet preset triggering conditions; and selecting the at least two photos to obtain target photos.

[0084] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0085] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic photo-based screening method, characterized in that, include: The acquired images are analyzed to obtain the analysis results; If the analysis results meet the preset triggering conditions, at least two photos will be taken; The target photo is obtained by filtering the at least two photos.

2. The automatic photo screening method according to claim 1, characterized in that, The analysis of the acquired images to obtain analysis results includes: The acquired image is analyzed from at least one first dimension to obtain the analysis result; the first dimension includes any one of the following: scene, object, and lighting.

3. The automatic photo screening method according to claim 2, characterized in that, The analysis of the acquired image from at least one first dimension to obtain the analysis result includes: An image analysis model is used to analyze the acquired image from at least one first dimension to obtain the analysis result.

4. The automatic photo screening method according to claim 1, characterized in that, The triggering conditions include at least one of the following: target scene, object features, lighting features, and composition logic.

5. The automatic photo screening method according to claim 1, characterized in that, The requirement to take at least two photos includes: Based on a preset shooting cycle, take at least two photos; or, Based on preset rules, at least two photos are taken.

6. The automatic photo screening method according to claim 1, characterized in that, The process of filtering the at least two photos to obtain the target photo includes: Based on a preset screening period, a large model is used to screen the at least two photos to obtain the target photo.

7. The automatic photo screening method according to claim 6, characterized in that, The process of using a large model to filter the at least two photos to obtain the target photo includes: For each of the at least two photos, a large model is used to analyze and score from at least one second dimension to obtain a score for each second dimension; The target score is determined based on the scores of each of the second dimensions; The photos corresponding to the top N target scores from each of the target scores are determined as the target photos.

8. The automatic photo screening method according to claim 7, characterized in that, The second dimension includes any of the following: aesthetic evaluation, content value, technical quality, character expressions, character actions, character emotions, and whether the scene is the same.

9. An automatic photo-screening device, characterized in that, include: The analysis module is used to analyze the acquired images and obtain the analysis results; The camera module is used to take at least two photos when the analysis results meet preset triggering conditions. The filtering module is used to filter the at least two photos to obtain the target photo.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the automatic photo screening method as described in any one of claims 1 to 8.

Images