Virtual shooting scene optimization method and system, electronic device and storage medium

Abstract

The application provides a virtual shooting scene optimization method and system, electronic equipment and a storage medium, wherein after obtaining a current shooting image, camera parameters and virtual shooting scene information, image feature recognition is performed on the current shooting image to obtain image feature information of a shooting subject in the current shooting image; then, a virtual level of the shooting subject is generated based on the shooting subject and the virtual shooting scene information; finally, the virtual shooting scene is optimized according to the camera parameters, the image feature information of the shooting subject and the virtual level. The application optimizes the virtual shooting scene dynamically, saves memory and effectively improves the picture frame rate, thereby solving the problem that the existing LED virtual shooting system needs to rely on high-power hardware and has high performance requirements.

Description

Technical Field

[0001] This invention relates to the field of virtual scene technology, and more specifically, to a method, system, electronic device, and storage medium for optimizing virtual shooting scenes. Background Technology

[0002] With the continuous development of technology, virtual scenes are widely used in various film and television industries. In a virtual scene, it is necessary to determine whether the current shot needs a complete and clear background and which part needs clear content. The demand here is determined and controlled by the director's control over the picture. Therefore, the artistic elements of the director's picture design and the focal length of the picture, as well as the rules of near, medium and far shots, are all based on biological visual simulation to guide the audience's view and story narration.

[0003] In existing technologies, LED virtual studios are used as the core, and a real-world fusion shooting is achieved by combining a real-time rendering engine (such as Unreal Engine) with a motion capture system (camera positioning). However, existing LED virtual shooting systems rely on high-power hardware and have high performance requirements. Summary of the Invention

[0004] In view of this, the present invention provides a virtual shooting scene optimization method, system, electronic device and storage medium to solve the problem that existing LED virtual shooting systems rely on high-power hardware and have high performance requirements.

[0005] The first aspect of this invention provides a method for optimizing virtual shooting scenes, comprising:

[0006] Acquire the currently captured image, camera parameters, and virtual shooting scene information;

[0007] Image feature recognition is performed on the currently captured image to obtain image feature information of the subject in the currently captured image;

[0008] Based on the shooting subject and the virtual shooting scene information, a virtual layer of the shooting subject is generated;

[0009] The virtual shooting scene is optimized based on the camera parameters, the image feature information of the subject being photographed, and the virtual hierarchy.

[0010] Optionally, the virtual shooting scene information includes at least the background of the virtual shooting scene and various objects in the virtual shooting scene, and the step of generating a virtual hierarchy of the shooting subject based on the shooting subject and the virtual shooting scene information includes:

[0011] A virtual model is created for the subject being photographed to obtain a virtual character, and the positional relationship between the virtual character and the background of the virtual shooting scene is determined.

[0012] Based on the positional relationship between the virtual character and the background of the virtual shooting scene, the virtual character and various objects in the virtual shooting scene are spatially divided to obtain the volumetric hierarchy of the shooting subject;

[0013] Based on the positional relationship between the virtual character and the background of the virtual shooting scene, a planar hierarchy is divided to obtain the planar hierarchy of the shooting subject.

[0014] Optionally, optimizing the virtual shooting scene based on the camera parameters, the image feature information of the subject being photographed, and the virtual hierarchy includes:

[0015] If there is a subject in the currently captured image, the first optimization strategy for the target is determined based on the framing range in the image feature information of the subject.

[0016] Based on the target, the first optimization strategy optimizes the virtual shooting scene according to the camera parameters and the image feature information of the shooting subject;

[0017] If there are at least two subjects in the current captured image, the priority of each subject is determined according to the virtual hierarchy of each subject.

[0018] The virtual shooting scene is optimized based on the camera parameters, the priority of each subject being photographed, and image feature information.

[0019] Optionally, if there are two subjects in the currently captured image, optimizing the virtual shooting scene based on the camera parameters, the image feature information of the subjects, and the virtual hierarchy includes:

[0020] If the framing range of the first subject is within the first framing range, the target subject is determined based on the first distance between the first subject and the second subject; wherein, the priority of the first subject is greater than the priority of the second subject.

[0021] A second optimization strategy is determined based on the framing range of the target subject, and the virtual shooting scene is optimized based on the second optimization strategy according to the camera parameters and the image feature information of the target subject.

[0022] If the framing range of the first shooting subject belongs to the second framing range, a target third optimization strategy is determined based on the framing range of the first shooting subject, and the virtual shooting scene is optimized based on the camera parameters and the image feature information of the first shooting subject according to the target third optimization strategy.

[0023] Optionally, the virtual shooting scene optimization method further includes:

[0024] Based on the motion states of the first and second shooting subjects, determine whether it is necessary to replace the first and second shooting subjects.

[0025] If necessary, the first shooting subject and the second shooting subject are replaced, and then the second optimization strategy step of determining the target based on the framing range of the target shooting subject is executed.

[0026] Optionally, if there are two or more subjects in the currently captured image, optimizing the virtual shooting scene based on the camera parameters, the image feature information of the subjects, and the virtual hierarchy includes:

[0027] If the framing range of the third shooting subject belongs to the first framing range, then the volume level and image feature information of the third shooting subject are adjusted to obtain the first volume level and first image feature information of the third shooting subject; wherein, the third shooting subject is the shooting subject with the highest priority among all the shooting subjects;

[0028] The fourth optimization strategy for the target is determined based on the framing range of the third shooting subject;

[0029] Based on the objective, the fourth optimization strategy optimizes the virtual shooting scene according to the camera parameters, the first volume level of the third shooting subject, and the first image feature information.

[0030] Optionally, the virtual shooting scene optimization method further includes:

[0031] If the framing range of the third shooting subject is within the second framing range, then the target fifth optimization strategy is determined based on the framing range of the third shooting subject, and the average value of the second distance between each shooting subject is calculated.

[0032] If the average value is greater than the second preset threshold, the volume level and image feature information of the third subject are adjusted to obtain the second volume level and second image feature information of the third subject.

[0033] Based on the fifth optimization strategy, the virtual shooting scene is optimized according to the camera parameters, the second volume level of the third shooting subject, and the second image feature information;

[0034] If the average value is not greater than the second preset threshold, then the virtual shooting scene is optimized based on the target fifth optimization strategy according to the camera parameters, the volume level of the third shooting subject, and image feature information.

[0035] A second aspect of the present invention provides a virtual shooting scene optimization system, comprising:

[0036] The acquisition unit is used to acquire the currently captured image, camera parameters, and virtual shooting scene information;

[0037] The feature recognition unit is used to perform image feature recognition on the currently captured image to obtain image feature information of the subject in the currently captured image;

[0038] A virtual hierarchy generation unit is used to generate a virtual hierarchy of the shooting subject based on the shooting subject and the virtual shooting scene information;

[0039] The first optimization unit is used to optimize the virtual shooting scene based on the camera parameters, the image feature information of the shooting subject, and the virtual hierarchy.

[0040] Optionally, the virtual shooting scene information includes at least the background of the virtual shooting scene and various objects in the virtual shooting scene, and the virtual hierarchy generation unit includes:

[0041] The positional relationship determination unit is used to perform virtual modeling of the shooting subject to obtain a virtual character, and to determine the positional relationship between the virtual character and the background of the virtual shooting scene;

[0042] The volume hierarchy determination unit is used to spatially divide the virtual character and various objects in the virtual shooting scene according to the positional relationship between the virtual character and the background of the virtual shooting scene, so as to obtain the volume hierarchy of the shooting subject.

[0043] The planar hierarchy determination unit is used to divide the planar hierarchy based on the positional relationship between the virtual character and the background of the virtual shooting scene, so as to obtain the planar hierarchy of the shooting subject.

[0044] Optionally, the first optimization unit includes:

[0045] The first optimization strategy determination unit is used to determine the first optimization strategy based on the framing range in the image feature information of the subject if there is a subject in the current captured image.

[0046] The second optimization unit is used to optimize the virtual shooting scene based on the target first optimization strategy according to the camera parameters and the image feature information of the shooting subject;

[0047] The subject priority determination unit is used to determine the priority of each subject based on the virtual hierarchy of each subject if there are at least two subjects in the current captured image.

[0048] The second optimization unit is further configured to optimize the virtual shooting scene based on the camera parameters, the priority of each of the shooting subjects, and image feature information.

[0049] Optionally, if there are two subjects in the currently captured image, the second optimization unit, when optimizing the virtual shooting scene based on the camera parameters, the priority of each subject, and image feature information, includes:

[0050] The target shooting subject determination unit is used to determine the target shooting subject based on the first distance between the first shooting subject and the second shooting subject if the framing range of the first shooting subject is within the first framing range; wherein the priority of the first shooting subject is greater than the priority of the second shooting subject.

[0051] The second optimization strategy determination unit is used to determine the second optimization strategy based on the framing range of the target subject, and optimize the virtual shooting scene based on the second optimization strategy according to the camera parameters and the image feature information of the target subject.

[0052] The target third optimization strategy determination unit is used to determine the target third optimization strategy based on the framing range of the first shooting subject if the framing range of the first shooting subject belongs to the second framing range, and optimize the virtual shooting scene based on the target third optimization strategy according to the camera parameters and the image feature information of the first shooting subject.

[0053] Optionally, the virtual shooting scene optimization system further includes:

[0054] The motion state acquisition unit is used to determine whether it is necessary to replace the first shooting subject with the second shooting subject based on the motion state of the first shooting subject and the second shooting subject;

[0055] The replacement unit is used to replace the first shooting subject with the second shooting subject if necessary, and then execute the step of determining the second optimization strategy based on the framing range of the target shooting subject.

[0056] Optionally, if there are two or more subjects in the currently captured image, the second optimization unit, when optimizing the virtual shooting scene based on the camera parameters, the image feature information of the subjects, and the virtual hierarchy, includes:

[0057] The first adjustment unit is used to adjust the volume level and image feature information of the third shooting subject if the framing range of the third shooting subject belongs to the first framing range, so as to obtain the first volume level and first image feature information of the third shooting subject; wherein, the third shooting subject is the shooting subject with the highest priority among all the shooting subjects;

[0058] The fourth optimization strategy determination unit is used to determine the fourth optimization strategy based on the framing range of the third shooting subject.

[0059] The third optimization subunit is used to optimize the virtual shooting scene based on the target fourth optimization strategy according to the camera parameters, the first volume level of the third shooting subject, and the first image feature information.

[0060] Optionally, the virtual shooting scene optimization system further includes:

[0061] The calculation unit is used to determine the target fifth optimization strategy based on the framing range of the third shooting subject if the framing range of the third shooting subject belongs to the second framing range, and to calculate the average value of the second distance between each shooting subject.

[0062] The second adjustment unit is used to adjust the volume level and image feature information of the third subject if the average value is greater than the second preset threshold, so as to obtain the second volume level and second image feature information of the third subject.

[0063] The third optimization subunit is also used to optimize the virtual shooting scene based on the target fifth optimization strategy according to the camera parameters, the second volume level of the third shooting subject, and the second image feature information;

[0064] The third optimization subunit is further configured to optimize the virtual shooting scene based on the target fifth optimization strategy according to the camera parameters, the volume hierarchy of the third shooting subject, and image feature information if the average value is not greater than the second preset threshold.

[0065] A third aspect of the present invention provides an electronic device, comprising:

[0066] One or more processors;

[0067] A storage device on which one or more programs are stored;

[0068] When the one or more programs are executed by the one or more processors, the one or more processors implement the virtual shooting scene optimization method as described in any one of the first aspects.

[0069] A fourth aspect of the present invention provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the virtual shooting scene optimization method as described in any one of the first aspects.

[0070] As can be seen from the above solutions, this invention provides a virtual shooting scene optimization method, system, electronic device, and storage medium. After acquiring the current shooting image, camera parameters, and virtual shooting scene information, image feature recognition is performed on the current shooting image to obtain the image feature information of the subject in the current shooting image. Then, based on the subject and virtual shooting scene information, a virtual layer of the subject is generated. Finally, the virtual shooting scene is optimized according to the camera parameters, the image feature information of the subject, and the virtual layer. This invention optimizes the virtual shooting scene dynamically, thereby saving memory and effectively improving the frame rate, thus solving the problem that existing LED virtual shooting systems rely on high-power hardware and have high performance requirements. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of the present 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 only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0072] Figure 1 A flowchart illustrating a virtual shooting scene optimization method provided in an embodiment of the present invention;

[0073] Figure 2 A schematic diagram of a framing range provided for another embodiment of the present invention;

[0074] Figure 3 A schematic diagram of a volumetric hierarchy provided for another embodiment of the present invention;

[0075] Figure 4 A schematic diagram of a planar hierarchy provided for another embodiment of the present invention;

[0076] Figure 5 A schematic diagram showing a face orientation according to another embodiment of the present invention;

[0077] Figure 6 A scene diagram showing two or more shooting subjects, provided as another embodiment of the present invention;

[0078] Figure 7 This is a schematic diagram of a virtual shooting scene optimization system provided in another embodiment of the present invention. Detailed Implementation

[0079] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0080] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0081] It should be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this invention are all information and data authorized by the user or fully authorized by all parties.

[0082] It should be noted that the concepts of "first" and "second" mentioned in this invention are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0083] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0084] This invention provides a method for optimizing virtual shooting scenes, such as... Figure 1 As shown, the specific steps include:

[0085] S101. Obtain the currently captured image, camera parameters, and virtual shooting scene information.

[0086] Among them, the camera parameters must include at least the actual camera focal length, which is not limited here.

[0087] It should be noted that the original host containing virtual assets cannot obtain the actual location of people or camera information in the captured footage. Therefore, in practical applications, this invention uses the image signal from a real camera and the current camera's focal length parameters, and connects them to a computer carrying the algorithm system via a signal stream.

[0088] S102. Perform image feature recognition on the currently captured image to obtain image feature information of the subject in the currently captured image.

[0089] Since there is a possibility that the captured image may not contain a subject, in the actual application of the present invention, before executing step S102, it is also possible to determine whether there is a subject in the current captured image. The determination method can be implemented by image recognition algorithms based on saliency detection, semantic segmentation, object detection, etc., and is not limited here.

[0090] Specifically, if it is determined that there is a subject in the current captured image, then step S102 is executed; if it is determined that there is no subject in the current captured image, then the virtual captured scene is optimized according to the mood type of the virtual captured scene.

[0091] The mood or atmosphere of the virtual shooting scene is set during the initial shooting and can be divided into I1 type (such as loneliness), I2 type (such as conflict), and I3 type (such as tranquility), without further limitation here.

[0092] When it is type I1, the way to optimize the virtual shooting scene is to reduce the scene color saturation. When it is type I2, the way to optimize the virtual shooting scene is to increase the resolution of the conflict area and reduce the resolution of the edge. When it is type I3, the way to optimize the virtual shooting scene is to reduce the scene animation rate.

[0093] In the actual application of this invention, reducing the color saturation of the scene, increasing the resolution of the conflict area, reducing the resolution of the edge perimeter, and reducing the scene animation rate can be a preset value, which is pre-set by experts, technicians, etc., or a corresponding option can be provided for control, which can be selected by the user. There is no limitation here.

[0094] The image feature information of the subject being photographed includes, but is not limited to: framing range C, eye direction guide FE, hand movement guide MH, body orientation FS, face orientation F, and subject movement state M, etc., which are not limited here.

[0095] In the actual application of this invention, the framing range C includes multiple ranges, such as C1 = close-up, C2 above the shoulders, C3 half-body, C4 full-body, C5 panoramic, and C6 wide-angle zoom.

[0096] Specifically, first locate C1, then gradually expand to obtain C2, C3, etc., such as Figure 2 As shown, no restrictions are imposed here.

[0097] Specifically, the method for locating C1 can be to first preset a target area that needs to be zoomed in on and tracked, such as a person's head or hands. There are no specific limitations here. Figure 2Taking the head as an example, C1 can be further expanded to obtain C2, C3, etc., based on the preset framing width, to increase the clarity of the displayed content.

[0098] In the practical application of this invention, C1 can be determined by image recognition algorithms, including but not limited to saliency detection, semantic segmentation, and object detection, etc., and no limitation is made here.

[0099] In the practical application of this invention, based on the obtained framing range, the acquisition of facial features of the character is further enhanced. For example, the positions of multiple marker points such as the facial features and ears of the virtual character will be determined with respect to the camera, thereby determining the face orientation F based on the marker points. This is not limited here. Similarly, the body orientation FS is obtained based on the marker points of the shoulders, chest, etc., to construct a human skeleton model and infer the frontal orientation of the body. This is not limited here.

[0100] In the practical application of this invention, the method of obtaining the eye orientation guide FE can be, but is not limited to, first detecting key facial points such as the eyeball, pupil, and corner of the eye, and then calculating the gaze vector through a three-dimensional face model to determine the gaze direction, i.e., the eye orientation guide FE.

[0101] In the practical application of this invention, the method of obtaining the hand motion guide MH may be, but is not limited to, determining the hand motion guide MH based on the trajectory of key points of the hand in continuous frames (such as the fingertip movement path), or analyzing the displacement direction and speed of pixels in the hand region in the time series to determine the hand motion guide MH. No limitation is made here.

[0102] In the actual application of this invention, the main motion state M can be obtained by comparing the differences between adjacent frame pixels, determining whether an object is moving, or calculating the motion vector field of each pixel, etc., and is not limited here.

[0103] It is understandable that it is not necessary to obtain all the image feature information of the subject being photographed; it can be one of the aforementioned image feature information, a combination thereof, etc., and there are no restrictions here.

[0104] S103. Based on the shooting subject and virtual shooting scene information, generate a virtual layer of the shooting subject.

[0105] The virtual layer includes a volumetric layer and a planar layer; the virtual shooting scene information includes at least the background of the virtual shooting scene and the various objects in the virtual shooting scene.

[0106] It should be noted that the volume layer in this invention refers to the position information formed by expanding a sphere outward from a 0.0.0 point set in the virtual shooting scene. Compared with the planar layer in the prior art, the planar layer in this invention adds position information determined by a 0.0.0 point set in the virtual shooting scene and the orientation of the virtual camera.

[0107] Optionally, in another embodiment of the present invention, one implementation of step S103 includes the following steps (steps A1 to A3):

[0108] Step A1: Create a virtual model of the subject to be photographed to obtain a virtual character, and determine the positional relationship between the virtual character and the background of the virtual shooting scene.

[0109] Understandably, since the focal length of the virtual camera (Vcam) needs to be consistent with that of the real camera, the image projected by the virtual camera is LED content in the real environment. There is a real-time synchronized mapping relationship between the virtual camera and the real camera. The positional relationship between the virtual character and the background of the virtual shooting scene, and between the real character and the LED background board, is a spatial consistency manifestation under a unified coordinate system.

[0110] Step A2: Based on the positional relationship between the virtual character and the background of the virtual shooting scene, divide the space of the virtual character and the various objects in the virtual shooting scene to obtain the volumetric hierarchy of the subject.

[0111] Specifically, such as Figure 3 As shown, firstly, a spherical phase diffusion is performed centered on the virtual character (P (0.0.0 point, currently represented by a cylinder in the attached diagram) (e.g., gradually diffusing outwards with a fixed radius), resulting in multiple volume levels (e.g. Figure 3 (X1, X2, X3, X4, X5, X6)

[0112] Then, determine the various objects in the virtual shooting scene (such as...) Figure 3 The relationship hierarchy between objects (a, b, c, d, e, f) and virtual characters in the virtual shooting scene is used to generate the volume hierarchy of the shooting subject based on the relationship hierarchy between all objects and virtual characters in the virtual shooting scene.

[0113] Taking object 'a' as an example, this section illustrates how to determine the hierarchical relationship between various objects and virtual characters in a virtual shooting scene. Figure 3 Object a is located between levels X2 and X3, and the relationship between this object and the virtual character is at level X2.

[0114] Step A3: Divide the planar layers based on the positional relationship between the virtual character and the background of the virtual shooting scene to obtain the planar layers of the shooting subject.

[0115] Specifically, the pose (position + attitude) of the real camera in physical space is acquired in real time through an optical / mechanical tracking system, and this pose is synchronously mapped onto the virtual camera (Vcam) in the virtual scene to ensure that the motion parameters of the virtual and real cameras are completely consistent. Then, as... Figure 4 As shown, the plane passing through the virtual character (P (0.0.0 point, currently represented by a cylinder in the attached diagram) and perpendicular to the optical axis of the virtual camera (Vcam) is taken as the depth reference plane Y0 where the virtual character is located. Based on this, a series of planes Y1, Y2, Y3, etc., parallel to Y0 are constructed sequentially along the line of sight into the depth of the scene (away from the virtual camera), or along the camera direction (closer to the virtual camera) to construct planes Y-1, Y-2, etc., parallel to Y0.

[0116] Each depth interval represents a planar layer. The difference in coordinates of foreground and background objects relative to Y0 in the depth direction is calculated based on the positional relationship between the virtual character and the background of the virtual shooting scene (i.e., the depth offset of each background object relative to the virtual character), and these objects are then assigned to their corresponding layers. This layer division is dynamically updated as the camera moves (i.e., the depth difference of each object relative to the virtual character is recalculated in real time) to ensure that the subject being filmed is always used as a reference.

[0117] Of course, in the practical application of this invention, the pose of the real camera in the virtual scene can also be calculated by inverse image calculation. Specifically, the distance from the camera to the object and the direction of the line of sight can be deduced first based on the known actual size of the reference object in the image (such as the height of a person, the distance between marker points) and its imaging pixel size, combined with the known true focal length or sensor size. If there are multiple reference points or multiple consecutive frames, the complete six-degree-of-freedom pose (position and attitude) of the camera can be solved by the PnP algorithm or bundle adjustment, which is not limited here.

[0118] S104. Optimize the virtual shooting scene based on camera parameters, image feature information of the subject being shot, and virtual layering.

[0119] Optionally, in another embodiment of the present invention, one implementation of step S104 includes the following steps (steps B1 to B5):

[0120] Step B1: Determine whether there is only one subject in the current captured image.

[0121] Specifically, if there is only one subject in the current captured image, proceed to step B2; if there are at least two subjects in the current captured image, proceed to step B4.

[0122] Step B2: Determine the target first optimization strategy based on the framing range in the image feature information of the subject being photographed.

[0123] Specifically, different optimization strategies are adopted based on different framing ranges as the primary optimization strategy.

[0124] Step B3: Based on the target-first optimization strategy, optimize the virtual shooting scene according to the camera parameters and the image feature information of the subject being shot.

[0125] Among them, the face orientation F can be divided into two types, such as Figure 5 As shown, one is horizontal (FF) and the other is vertical (FU); FF represents the face orientation (left and right direction), 180≥FF≥0 with the camera direction as 0°, and then normalized, 1≥FF≥0, 0 is facing the camera directly, 1 is facing away from the camera; similarly, 180≥FU≥0° with the camera direction as 90°, and then normalized: 1≥FU≥0, 0 is looking down, 1 is looking up, 0.5 is looking directly in front.

[0126] In this case, the body faces FS, 180≥FS≥0°, the horizontal direction is 0° with the camera direction as the reference, and the normalization processing is 1≥FS≥0, where 1 represents the front and back.

[0127] Continuing with the above example, when the framing range is C1 / C2, the corresponding optimization strategy (target-first optimization strategy) is: based on the changes in the values ​​of FF and FS and the FO value (the focal length of the actual camera) (ΔV). FO Adjust the V value.

[0128] It should be noted that the value of V in this invention affects the range of spatial and planar partitioning. When V=0, the range is infinitely large, indicating that no objects are compressed. When V is greater than or equal to a threshold (e.g., 0.9), it means that all objects will be affected. The setting is: 1≥V≥0. Its main purpose is to determine the scope of optimization.

[0129] Specifically, by default, the setting value V set It can be equal to 0.9, depending on the set value V. set and ΔV FO Adjust the V value; when both FF and FS are less than 0.1, set the character to face away from the camera, then set the V value. set =0, based on the set value V set and ΔV FO Adjust the V value; if both FF and FS are greater than or equal to 0.1, set the value V. set It equals 0.9, and then, according to the set value V set ΔV FO The variables FU, FE, MH, M, and their corresponding weights are adjusted to V values, which can be achieved using the following formula:

[0130] ;

[0131] in, These are base values, preset and modified by experts and technicians; no restrictions are imposed here. These are variables, such as FU, FE, MH, M, etc., which are not limited here. For variables The corresponding weights are preset and modified by experts and technicians, and are not limited here.

[0132] When the framing range is C3, the primary optimization strategy is: when both FF and FS are less than 0.1, the character is set to face away from the camera, with a set value of V. set It can be equal to 0.2, and then, according to the set value V set and ΔV FO Adjust the V value. Set the value V when both FF and FS are greater than or equal to 0.1. set Equals 0.9, combined with ΔV FO In addition, variables such as FU, FE, MH, and M, along with their corresponding weights, are added to adjust the V value. This can be achieved using the following formula:

[0133] ;

[0134] in, The weights corresponding to FU The weights corresponding to FE. The weights corresponding to MH, The weights corresponding to M are preset and modified by experts, technicians, etc., and are not limited here.

[0135] When the framing range is C4, the primary optimization strategy is: when both FF and FS are less than 0.1, the character is set to face away from the camera, with a set value of V. set It can be equal to 0.2, by... Change to and will Change to Increase the weight ratio of M and MH to increase the accuracy of the VM, VF, and VL regions. and This is pre-set and modified by experts and technicians; no restrictions are imposed here. Then, it is combined with ΔV. FO Adjust the V value. When both FF and FS are greater than or equal to 0.1, set it to a front-facing camera for the character, and set the V value accordingly. set It can be equal to 0.9, then, combined with ΔV FOIn addition, variables such as FU, FE, MH, and M, along with their corresponding weights, are added to adjust the V value. Furthermore, the weight corresponding to FU is adjusted based on the current value of FU. For example, when FU ≤ 0.25, the weight corresponding to FU is increased, and the judgment is made to look towards the ground. This can be achieved using the following formula:

[0136] ;

[0137] in, As the first FU weight, The second FU weight can be preset and changed by experts, technicians, etc., and is not limited here.

[0138] When the framing range is C5, the primary optimization strategy is: when both FF and FS are less than 0.1, the character is set to face away from the camera, with a set value of V. set It can be 0.1, by... Change to and will Change to Increase the weight ratio of M and MH to increase the accuracy of the VM, VF, and VL regions. and Pre-set and modified by experts and technicians, etc., are not limited here. When both FF and FS are greater than or equal to 0.1, it is set as a front-facing camera, with a setting value V. set It can be equal to 0.9, combined with ΔV FO In addition, variables such as FU, FE, MH, and M, along with their corresponding weights, are added to adjust the V value. This can be achieved using the following formula:

[0139] ;

[0140] When the framing range is C6, the primary optimization strategy is: without determining direction, set to artistic mode, and then... Change to and will Change to Increase the weight ratio of M and MH, combined with ΔV FO The adjustment of artistic conception and the adjustment of artistic conception weight V value can be achieved using the following formula:

[0141] ;

[0142] ;

[0143] in, This is an adjustment option for mood / environment; -s indicates a reduction in saturation, and +r indicates an increase in resolution of conflict areas. This refers to a reduced animation speed. For the pre-set artistic conception weight, These settings are pre-set and modified by experts and technicians, and no restrictions are imposed here.

[0144] Step B4: Determine the priority of each subject based on its virtual layer.

[0145] In the practical application of this invention, the priority of each shooting subject is mainly determined by the planar layer Y, such that the closer the subject is to Y0, the higher the priority.

[0146] However, since the LED screen in the real world displays a two-dimensional content, but is also a three-dimensional world, the present invention can further assist in positioning and detail processing by capturing the volumetric layer X of the subject.

[0147] It should be noted that X determines the five directions of up, down, left, right and forward, while Y determines the objects in the back direction (since the background occupies the most space, it is the main control, with X as an auxiliary factor). No restrictions are imposed here.

[0148] Step B5: Optimize the virtual shooting scene based on camera parameters, the priority of each subject, and image feature information.

[0149] Optionally, in another embodiment of the present invention, if there are two subjects in the currently captured image, then one implementation of step B5 includes the following steps (steps D1 to D5):

[0150] Step D1: Determine whether the framing range of the first subject belongs to the first framing range or the second framing range.

[0151] The image feature information of the first subject includes at least the field of view of the first subject.

[0152] Continuing with the above examples, the framing range includes C1 (close-up), C2 (shoulder and above), C3 (half-body), C4 (full-body), C5 (panoramic), and C6 (wide-angle zoom). In practical applications of this invention, the first framing range can be set to include C1, C2, and C3, and the second framing range can be set to include C4, C5, and C6; no limitation is made here.

[0153] Specifically, if it is determined that the framing range of the first subject belongs to the first framing range, then proceed to step D2; if it is determined that the framing range of the first subject belongs to the second framing range, then proceed to step D5.

[0154] Step D2: Calculate the first distance between the first subject and the second subject.

[0155] Among them, the priority of the first subject being photographed is higher than that of the second subject being photographed.

[0156] In the actual application of the present invention, the first distance between the first shooting subject and the second shooting subject can be calculated by the coordinates of the two shooting subjects in three-dimensional space, or by calculating the precise depth and horizontal position obtained by inverse calculation of the camera parameters, or by estimating the distance range of the two shooting subjects in the depth direction using the plane layer where the two shooting subjects are located, etc., which are not limited here.

[0157] Step D3: Determine whether the first distance is greater than the first preset threshold.

[0158] Specifically, if it is determined that the first distance is greater than the first preset threshold, then the first shooting subject and the second shooting subject are respectively taken as the target shooting subject, and step D4 is executed. If it is determined that the first distance is not greater than the first preset threshold, then the first shooting subject or the second shooting subject is selected as a target shooting subject, and step D4 is executed. It can be understood that since the priority of the first shooting subject is greater than that of the second shooting subject, the better solution is to select the first shooting subject as the target shooting subject.

[0159] Step D4: Determine the second optimization strategy for the target subject based on the framing range of the target subject, and optimize the virtual shooting scene based on the camera parameters and image feature information of the target subject according to the second optimization strategy.

[0160] The second optimization strategy is the optimization strategy corresponding to C1 / C2 or C3 in the above example. The specific implementation of optimizing the virtual shooting scene based on the second optimization strategy according to the camera parameters and the image feature information of the target subject can be referred to the implementation in step B3, which will not be repeated here.

[0161] Step D5: Determine the target third optimization strategy based on the framing range of the first subject, and optimize the virtual shooting scene based on the target third optimization strategy according to the camera parameters and the image feature information of the first subject.

[0162] The target third optimization strategy is the optimization strategy corresponding to C4, C5, or C6 in the above example. The specific implementation method for optimizing the virtual shooting scene based on the target third optimization strategy according to the camera parameters and the image feature information of the first shooting subject can be referred to the implementation method in step B3, which will not be repeated here.

[0163] It should be noted that when C=4 / 5 / 6, the X1 region must contain two subjects (that is, by adjusting the radius to create a sphere that expands outwards, the X1 sphere simultaneously contains both the first and second subjects), and the X1 region must always remain sharp. Specifically, the texture and model precision in this region can be optimized to ensure it remains sharp at all times.

[0164] Optionally, in another embodiment of the present invention, if there are two or more subjects in the currently captured image, such as... Figure 6 As shown, one implementation of step B5 includes the following steps (steps E1 to E9):

[0165] Step E1: Determine whether the framing range of the third subject belongs to the first framing range or the second framing range.

[0166] Among them, the third subject is the subject with the highest priority among all subjects.

[0167] Continuing with the above examples, the framing range includes C1 (close-up), C2 (shoulder and above), C3 (half-body), C4 (full-body), C5 (panoramic), and C6 (wide-angle zoom). In practical applications of this invention, the first framing range can be set to include C1, C2, and C3, and the second framing range can be set to include C4, C5, and C6; no limitation is made here.

[0168] Specifically, if it is determined that the framing range of the third subject belongs to the first framing range, then step E2 is executed; if it is determined that the framing range of the third subject belongs to the second framing range, then step E5 is executed.

[0169] Step E2: Adjust the volume level and image feature information of the third subject to obtain the first volume level and first image feature information of the third subject.

[0170] Specifically, the volume level and image feature information of the third subject are adjusted according to the first parameter adjustment information based on the number of subjects being photographed, to obtain the first volume level and first image feature information of the third subject.

[0171] The first parameter adjustment information includes the correspondence between the number of subjects being photographed, the range weight of the volume level, and the weight of M.

[0172] In the practical application of this invention, when the framing range is within the first framing range, i.e., C=1 / 2 / 3, distance is not considered, and it is set as a single subject (the image cannot accommodate other people; if there are, they must be very close). Then, based on the number of subjects being photographed, the range weight of the volume levels (such as X1, X2, etc.) is adjusted. The adjusted X1 range is calculated based on the original range of X1 and the adjusted range weight of X1, and the adjusted X2 range is calculated based on the original range of X2 and the adjusted range weight of X2, which is the first volume level. The weight of M is adjusted according to the number of subjects being photographed, and the adjusted M value is generated based on the original M value and the adjusted M weight. The first image feature information includes the adjusted M value and other unadjusted feature information (such as framing range C, eye direction guide FE, hand movement guide MH, body direction FS, face direction F, etc.).

[0173] Specifically, as the number of subjects increases, the amount of content in the frame (people and scenery) increases. Therefore, X1-X2 (scene) and M (probability of person movement) should both increase sequentially. Thus, the weights of M and X's range are adjusted synchronously based on the increase in the number of subjects. This can be a pre-defined correspondence; for example, when the number of subjects is 4, the range weight of X1 is... When the number of subjects being photographed is 6, the range weight of X1 is: , Greater than No restrictions are imposed here. The weight of M is similar and will not be elaborated further here.

[0174] Of course, as the framing range increases (C1 to C3), the content in the frame will also increase (people and scenes). Similarly, the range weight of X and the weight of M can be further adjusted according to the change in the framing range, without any restrictions here.

[0175] Step E3: Determine the target fourth optimization strategy based on the framing range of the third shooting subject.

[0176] The fourth optimization strategy is the optimization strategy corresponding to C1 / C2 or C3 in the above example. The corresponding optimization strategy is selected based on the framing range of the third shooting subject, and is used as the fourth optimization strategy.

[0177] Step E4: Based on the target, the fourth optimization strategy optimizes the virtual shooting scene according to the camera parameters, the first volume level of the third shooting subject, and the first image feature information.

[0178] Specifically, firstly, a condition is added: the first volume level (such as the adjusted X1 region) is always kept clear, and the virtual shooting scene is optimized based on the target fourth optimization strategy according to the camera parameters and the first image feature information. The specific implementation of the optimization can be referred to the implementation in step B3, which will not be repeated here.

[0179] Step E5: Determine the fifth optimization strategy for the target based on the framing range of the third shooting subject, and calculate the average value of the second distance between each shooting subject.

[0180] Among them, the fifth optimization strategy is the optimization strategy corresponding to C4, C5, or C6 in the above example.

[0181] In the actual application of this invention, the second distance between each shooting subject can be calculated by the coordinates of the two shooting subjects in three-dimensional space, or by calculating the precise depth and horizontal position obtained by inverse calculation of the camera parameters, or by estimating the distance range of the two shooting subjects in the depth direction using the plane layer where the two shooting subjects are located, etc., which are not limited here.

[0182] Step E6: Determine whether the average value is greater than the second preset threshold.

[0183] Specifically, if the average value is determined to be greater than the second preset threshold, then step E7 is executed; if the average value is determined to be less than the second preset threshold, then step E9 is executed.

[0184] Step E7: Adjust the volume level and image feature information of the third subject to obtain the second volume level and second image feature information of the third subject.

[0185] Specifically, based on the number of subjects being photographed, the volume level and image feature information of the third subject are adjusted according to the second parameter adjustment information to obtain the second volume level and second image feature information of the third subject.

[0186] The second parameter adjustment information includes the correspondence between the number of subjects being photographed, the range weight of the volume level, the weight of MH, and the weight of M.

[0187] In the practical application of this invention, when the framing range belongs to the second framing range, i.e., C=4 / 5 / 6, the range weight of the expanded volume level (such as X1, X2, etc.) is adjusted according to the number of subjects being photographed. The adjusted X1 range is calculated based on the original range of X1 and the adjusted range weight of X1, and the adjusted X2 range is calculated based on the original range of X2 and the adjusted range weight of X2, which is the second volume level. The weight of M is adjusted according to the number of subjects being photographed, and the adjusted M value is generated based on the original M value and the adjusted M weight. The weight of MH is adjusted according to the number of subjects being photographed, and the adjusted MH value is generated based on the original MH value and the adjusted MH weight. The second image feature information includes the adjusted MH value, the adjusted M value, and other unadjusted feature information (such as framing range C, eye direction guide FE, body direction FS, face direction F, etc.).

[0188] The method for adjusting the weight of MH can refer to the method for adjusting the weight of M in the above embodiment, and will not be repeated here.

[0189] It's important to note that M and MH both refer to motion attributes. Stronger M and MH values ​​will influence the mood or atmosphere of the current scene, for example, switching from a serene, scenic mood to a dynamic, action-packed mood. The main objects are the VM (moving objects in the virtual shooting scene, such as destroyed objects) and VF (special effects in the virtual shooting scene). These are considered to actively respond to changes in the shooting, increasing the detail of objects. Simultaneously, the VL (regional brightness in the virtual shooting scene) of the corresponding area during the action also changes accordingly, thus focusing the visual center.

[0190] Step E8: Based on the fifth optimization strategy of the objective, optimize the virtual shooting scene according to the camera parameters, the second volume level of the third shooting subject and the second image feature information.

[0191] Specifically, firstly, a condition is added: the second volume level (such as the adjusted X1 region) is always kept clear, and the virtual shooting scene is optimized based on the target fourth optimization strategy according to the camera parameters and the second image feature information. The specific implementation of the optimization can be referred to the implementation in step B3, which will not be repeated here.

[0192] Step E9: Based on the fifth optimization strategy, optimize the virtual shooting scene according to camera parameters, the volume hierarchy of the third shooting subject, and image feature information.

[0193] For specific details on the optimized implementation method, please refer to the implementation method in step B3, which will not be repeated here.

[0194] Optionally, in another embodiment of the present invention, one implementation of the virtual shooting scene optimization method further includes the following steps (steps F1 to F2):

[0195] Step F1: Based on the motion states of the first and second shooting subjects, determine whether it is necessary to replace the first and second shooting subjects.

[0196] In the practical application of this invention, the distance between the subject and the background LED can be used as the main reference factor to determine whether the first shooting subject and the second shooting subject need to be replaced. At the same time, the comprehensive adaptability of motion trajectory matching, visual coherence and production efficiency can also be considered, but no limitation is made here.

[0197] Step F2: If it is necessary to replace the first subject with the second subject, replace the first subject with the second subject.

[0198] It should be noted that if the initial priority of the first subject is higher than that of the second subject, after replacement, the priority of the second subject becomes higher than that of the first subject. Then, the step of determining the target second optimization strategy based on the image feature information of the first subject (after replacement) is executed.

[0199] As can be seen from the above scheme, the present invention provides a virtual shooting scene optimization method. After acquiring the current shooting image, camera parameters, and virtual shooting scene information, image feature recognition is performed on the current shooting image to obtain the image feature information of the subject in the current shooting image. Then, based on the subject and virtual shooting scene information, a virtual layer of the subject is generated. Finally, the virtual shooting scene is optimized according to the camera parameters, the image feature information of the subject, and the virtual layer. The present invention optimizes the virtual shooting scene dynamically, thereby saving memory and effectively improving the frame rate, thus solving the problem that existing LED virtual shooting systems rely on high-power hardware and have high performance requirements.

[0200] Another embodiment of the present invention provides a virtual shooting scene optimization system, such as... Figure 7 As shown, it specifically includes:

[0201] The acquisition unit 701 is used to acquire the currently captured image, camera parameters, and virtual shooting scene information.

[0202] The feature recognition unit 702 is used to perform image feature recognition on the currently captured image to obtain image feature information of the subject in the currently captured image.

[0203] The virtual layer generation unit 703 is used to generate a virtual layer of the subject based on the subject being photographed and the virtual shooting scene information.

[0204] The first optimization unit 704 is used to optimize the virtual shooting scene based on camera parameters, image feature information of the subject being shot, and virtual hierarchy.

[0205] For details on the specific operation of the units disclosed in the above embodiments of the present invention, please refer to the corresponding method embodiments, such as... Figure 1 As shown, it will not be elaborated further here.

[0206] As can be seen from the above scheme, the present invention provides a virtual shooting scene optimization system. After acquiring the current shooting image, camera parameters, and virtual shooting scene information, image feature recognition is performed on the current shooting image to obtain the image feature information of the subject in the current shooting image. Then, based on the subject and virtual shooting scene information, a virtual layer of the subject is generated. Finally, the virtual shooting scene is optimized according to the camera parameters, the image feature information of the subject, and the virtual layer. The present invention optimizes the virtual shooting scene dynamically, thereby saving memory and effectively improving the frame rate, thus solving the problem that existing LED virtual shooting systems rely on high-power hardware and have high performance requirements.

[0207] Another embodiment of the present invention provides an electronic device, comprising:

[0208] One or more processors.

[0209] A storage device on which one or more programs are stored.

[0210] When the one or more programs are executed by the one or more processors, the one or more processors implement the virtual shooting scene optimization method as described in the above embodiments.

[0211] Another embodiment of the present invention provides a storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, implements the virtual shooting scene optimization method as described in the above embodiments.

[0212] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0213] It should be noted that the computer-readable medium described above in this invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0214] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0215] Another embodiment of the present invention provides a computer program product, which, when executed, is used to perform the above-described virtual shooting scene optimization method.

[0216] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, it performs the functions defined in the methods of the embodiments of the present invention.

[0217] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in this invention is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely exemplary forms for implementing the invention.

[0218] While several specific implementation details are included in the foregoing discussion, these should not be construed as limiting the scope of the invention. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0219] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention is not limited to the specific combination of the above-described technical features, but also includes other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with technical features having similar functions in the present invention.

Claims

1. A method for optimizing virtual shooting scenes, characterized in that, include: Acquire the currently captured image, camera parameters, and virtual shooting scene information; Image feature recognition is performed on the currently captured image to obtain image feature information of the subject in the currently captured image; Based on the shooting subject and the virtual shooting scene information, a virtual layer of the shooting subject is generated; The virtual shooting scene is optimized based on the camera parameters, the image feature information of the subject being photographed, and the virtual hierarchy.

2. The virtual shooting scene optimization method according to claim 1, characterized in that, The virtual shooting scene information includes at least the background of the virtual shooting scene and various objects in the virtual shooting scene. The step of generating a virtual hierarchy for the shooting subject based on the shooting subject and the virtual shooting scene information includes: A virtual model is created for the subject being photographed to obtain a virtual character, and the positional relationship between the virtual character and the background of the virtual shooting scene is determined. Based on the positional relationship between the virtual character and the background of the virtual shooting scene, the virtual character and various objects in the virtual shooting scene are spatially divided to obtain the volumetric hierarchy of the shooting subject; Based on the positional relationship between the virtual character and the background of the virtual shooting scene, a planar hierarchy is divided to obtain the planar hierarchy of the shooting subject.

3. The virtual shooting scene optimization method according to claim 1, characterized in that, The optimization of the virtual shooting scene based on the camera parameters, the image feature information of the subject being photographed, and the virtual hierarchy includes: If there is a subject in the currently captured image, the first optimization strategy for the target is determined based on the framing range in the image feature information of the subject. Based on the target, the first optimization strategy optimizes the virtual shooting scene according to the camera parameters and the image feature information of the shooting subject; If there are at least two subjects in the current captured image, the priority of each subject is determined according to the virtual hierarchy of each subject. The virtual shooting scene is optimized based on the camera parameters, the priority of each subject being photographed, and image feature information.

4. The virtual shooting scene optimization method according to claim 3, characterized in that, If there are two subjects in the currently captured image, the optimization of the virtual shooting scene based on the camera parameters, the image feature information of the subjects, and the virtual hierarchy includes: If the framing range of the first subject is within the first framing range, the target subject is determined based on the first distance between the first subject and the second subject; wherein, the priority of the first subject is greater than the priority of the second subject. A second optimization strategy is determined based on the framing range of the target subject, and the virtual shooting scene is optimized based on the second optimization strategy according to the camera parameters and the image feature information of the target subject. If the framing range of the first shooting subject belongs to the second framing range, a target third optimization strategy is determined based on the framing range of the first shooting subject, and the virtual shooting scene is optimized based on the target third optimization strategy according to the camera parameters and the image feature information of the first shooting subject.

5. The virtual shooting scene optimization method according to claim 4, characterized in that, Also includes: Based on the motion states of the first and second shooting subjects, determine whether it is necessary to replace the first and second shooting subjects. If necessary, the first shooting subject and the second shooting subject are replaced, and then the second optimization strategy step of determining the target based on the framing range of the target shooting subject is executed.

6. The virtual shooting scene optimization method according to claim 4, characterized in that, If there are two or more subjects in the currently captured image, the optimization of the virtual shooting scene based on the camera parameters, the image feature information of the subjects, and the virtual hierarchy includes: If the framing range of the third shooting subject belongs to the first framing range, then the volume level and image feature information of the third shooting subject are adjusted to obtain the first volume level and first image feature information of the third shooting subject; wherein, the third shooting subject is the shooting subject with the highest priority among all the shooting subjects; The fourth optimization strategy for the target is determined based on the framing range of the third shooting subject; Based on the objective, the fourth optimization strategy optimizes the virtual shooting scene according to the camera parameters, the first volume level of the third shooting subject, and the first image feature information.

7. The virtual shooting scene optimization method according to claim 6, characterized in that, Also includes: If the framing range of the third shooting subject is within the second framing range, then the target fifth optimization strategy is determined based on the framing range of the third shooting subject, and the average value of the second distance between each shooting subject is calculated. If the average value is greater than the second preset threshold, the volume level and image feature information of the third subject are adjusted to obtain the second volume level and second image feature information of the third subject. Based on the fifth optimization strategy, the virtual shooting scene is optimized according to the camera parameters, the second volume level of the third shooting subject, and the second image feature information; If the average value is not greater than the second preset threshold, then the virtual shooting scene is optimized based on the target fifth optimization strategy according to the camera parameters, the volume level of the third shooting subject, and image feature information.

8. A virtual shooting scene optimization system, characterized in that, include: The acquisition unit is used to acquire the currently captured image, camera parameters, and virtual shooting scene information; The feature recognition unit is used to perform image feature recognition on the currently captured image to obtain image feature information of the subject in the currently captured image; A virtual hierarchy generation unit is used to generate a virtual hierarchy of the shooting subject based on the shooting subject and the virtual shooting scene information; The first optimization unit is used to optimize the virtual shooting scene based on the camera parameters, the image feature information of the shooting subject, and the virtual hierarchy.

9. An electronic device, characterized in that, include: A processor and a memory, wherein the processor is configured to call and execute a program stored in the memory; The memory is used to store a program for implementing the virtual shooting scene optimization method as described in any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores computer-executable instructions for performing the virtual shooting scene optimization method as described in any one of claims 1-7.

Images