A detail level model detection method and device

Abstract

The present disclosure provides a detail level model detection method and device, a standard model and a detail level model of a target virtual object are obtained; the standard model and the detail level model are preprocessed to obtain a first observation image of the standard model in an observation coordinate system matched with the detail level model, and a second observation image of the detail level model in the observation coordinate system; a target difference degree between the first observation image and the second observation image is determined based on geometric difference features and / or attribute difference features between the first observation image and the second observation image; and in a case where the target difference degree is greater than or equal to a qualified threshold, it is determined that the detail level model fails the quality detection. The embodiments of the present disclosure can make the target difference degree more accurate, thereby improving the accuracy of the quality detection and facilitating the improvement of the quality of the detail level model.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer, and in particular, to a levels of detail model detection method and device. BACKGROUND

[0002] In a virtual three-dimensional scene, various virtual models need to be displayed, and some virtual models need to be deployed with one or more levels of detail (LOD) models. LOD model is a technology in the field of image display, which considers that when an object covers a small area of a screen, a relatively rough model with a low number of faces can be used to describe the object, and a geometric level model for visible face determination algorithm is given to quickly draw a complex scene.

[0003] For the construction of the LOD model, it is difficult to balance the calculation performance and the visual effect of the model due to the complex structure of the LOD model, and it is difficult to accurately detect the LOD model, resulting in the difficulty of optimizing the LOD model. SUMMARY

[0004] The embodiments of the present disclosure at least provide a levels of detail model detection method and device.

[0005] In a first aspect, the embodiments of the present disclosure provide a levels of detail model detection method, comprising:

[0006] obtaining a standard model and a levels of detail model of a target virtual object;

[0007] preprocessing the standard model and the levels of detail model to obtain a first observation image of the standard model in an observation coordinate system matched with the levels of detail model, and a second observation image of the levels of detail model in the observation coordinate system;

[0008] determining a target difference degree between the first observation image and the second observation image based on geometric difference features and / or attribute difference features between the first observation image and the second observation image;

[0009] in a case where the target difference degree is greater than or equal to a qualified threshold, determining that the levels of detail model fails the quality detection.

[0010] In an optional implementation, the geometric difference features between the first observation image and the second observation image include first geometric difference features of each first plane vertex in the first observation image, and second geometric difference features of each second plane vertex in the second observation image.

[0011] The geometric difference features between the first observation image and the second observation image are obtained by the following steps:

[0012] adjusting positions of the first observation image and the second observation image in the observation coordinate system so that center points of the first observation image and the second observation image coincide;

[0013] determining a first distance from each first plane vertex in the first observation image to a respective second plane in the second observation image; for each first plane vertex in the first observation image, taking the smallest first distance as a first geometric difference feature of the first plane vertex; and,

[0014] determining a second distance from each second plane vertex in the second observation image to a respective first plane in the first observation image; for each second plane vertex in the second observation image, taking the smallest second distance as a second geometric difference feature of the second plane vertex.

[0015] In an optional implementation, the target difference degree between the first observation image and the second observation image is determined based on the geometric difference features between the first observation image and the second observation image, including:

[0016] determining a target geometric difference feature as the maximum of the first geometric difference features and the second geometric difference features;

[0017] determining the target difference degree between the first observation image and the second observation image based on a first size corresponding to a coordinate range of any pixel point in the first observation image or the second observation image in the world coordinate system, and the target geometric difference feature.

[0018] In an optional implementation, the first size corresponding to the coordinate range of any pixel point in the first observation image or the second observation image in the world coordinate system is determined by:

[0019] obtaining a second size of any pixel point in the first observation image or the second observation image, a picture height of the first observation image or the second observation image, a shooting angle parameter of a camera when preprocessing the standard model or the detail level model, and a third distance between the camera and the standard model or the detail level model;

[0020] determining the first size based on a ratio between the second size, the shooting angle parameter, and the third distance and the picture height.

[0021] In an optional implementation, the attribute difference feature between the first observation image and the second observation image is obtained by:

[0022] convert the first observation image and the second observation image into voxels to obtain a first mesh model corresponding to the first observation image and a second mesh model corresponding to the second observation image;

[0023] For each key point, determine a first plane in which the key point is located in the first mesh model, and a second plane in which the key point is located in the second mesh model;

[0024] Based on the attribute information of the first mesh matched with each first plane vertex of the first plane, determine the attribute information of the key point in the first observation image;

[0025] Based on the attribute information of the second mesh matched with each second plane vertex of the second plane, determine the attribute information of the key point in the second observation image;

[0026] Determine the target ratio between the attribute information of the key point in the first observation image and the attribute information of the key point in the second observation image, and take the target ratio of each key point as the attribute difference feature between the first observation image and the second observation image.

[0027] In an optional implementation, based on the attribute difference feature between the first observation image and the second observation image, determining the target difference degree between the first observation image and the second observation image, comprises:

[0028] For each key point, based on the target ratio of the key point, determine the attribute difference weight of the key point;

[0029] Based on the attribute difference weight of each key point, filter a plurality of adjacent target key points from the plurality of key points, so that the difference evaluation value of the cube composed of the target key points is maximum, and the difference evaluation value is determined based on the attribute difference weight of each key point in the cube;

[0030] Based on the ratio between the median of the side length of the cube composed of the target key points and the first size of any pixel point in the world coordinate system in the first observation image or the second observation image, determine the target difference degree between the first observation image and the second observation image.

[0031] In an optional implementation, based on the geometric difference feature and the attribute difference feature between the first observation image and the second observation image, determining the target difference degree between the first observation image and the second observation image, comprises:

[0032] Based on the geometric difference feature, determine the first difference degree between the first observation image and the second observation image;

[0033] determine a second difference degree between the first observation image and the second observation image based on the attribute difference feature;

[0034] determine a target difference degree between the first observation image and the second observation image based on the first difference degree and the second difference degree.

[0035] In a second aspect, the embodiments of the present disclosure further provide a detail level model detection device, comprising:

[0036] an acquisition module configured to acquire a standard model and a detail level model of a target virtual object;

[0037] a preprocessing module configured to pre-process the standard model and the detail level model to obtain a first observation image of the standard model in an observation coordinate system matched with the detail level model, and a second observation image of the detail level model in the observation coordinate system;

[0038] a detection module configured to determine a target difference degree between the first observation image and the second observation image based on a geometric difference feature and / or an attribute difference feature between the first observation image and the second observation image;

[0039] a judgment module configured to determine that the detail level model fails the quality detection if the target difference degree is greater than or equal to a qualified threshold.

[0040] In an optional implementation, the geometric difference feature between the first observation image and the second observation image comprises a first geometric difference feature of each first plane vertex in the first observation image, and a second geometric difference feature of each second plane vertex in the second observation image.

[0041] When acquiring the geometric difference feature between the first observation image and the second observation image, the acquisition module is configured to:

[0042] adjust positions of the first observation image and the second observation image in the observation coordinate system so that center points of the first observation image and the second observation image coincide;

[0043] determine a first distance from each first plane vertex in the first observation image to each second plane in the second observation image; for each first plane vertex in the first observation image, take the smallest first distance as the first geometric difference feature of the first plane vertex; and

[0044] determining a second distance of each second plane vertex in the second observation image to a respective first plane in the first observation image; and taking the smallest second distance as a second geometric difference feature of the second plane vertex for each second plane vertex in the second observation image.

[0045] In an optional implementation, when determining the target difference degree between the first observation image and the second observation image based on the geometric difference features between the first observation image and the second observation image, the detection module is configured to:

[0046] determining a target geometric difference feature from the first geometric difference features and the second geometric difference features;

[0047] determining a target difference degree between the first observation image and the second observation image based on a first size corresponding to a coordinate range of any pixel in the first observation image or the second observation image in a world coordinate system, and the target geometric difference feature.

[0048] In an optional implementation, when determining the first size corresponding to the coordinate range of any pixel in the first observation image or the second observation image in the world coordinate system, the detection module is configured to:

[0049] obtaining a second size of any pixel in the first observation image or the second observation image, a picture height of the first observation image or the second observation image, a shooting angle parameter of a camera when pre-processing the standard model or the level-of-detail model, and a third distance between the camera and the standard model or the level-of-detail model;

[0050] determining the first size based on a ratio between the second size, the shooting angle parameter, and the third distance and the picture height.

[0051] In an optional implementation, when obtaining the attribute difference feature between the first observation image and the second observation image, the detection module is configured to:

[0052] converting the first observation image and the second observation image into voxels to obtain a first mesh model corresponding to the first observation image and a second mesh model corresponding to the second observation image;

[0053] for each key point, determining a first plane in which the key point is located in the first mesh model, and a second plane in which the key point is located in the second mesh model;

[0054] determine attribute information of the key points in the first observation image based on attribute information of a first mesh matched with each first plane vertex of the first plane;

[0055] determine attribute information of the key points in the second observation image based on attribute information of a second mesh matched with each second plane vertex of the second plane;

[0056] determine a target ratio between the attribute information of the key points in the first observation image and the attribute information of the key points in the second observation image, and take the target ratio of each key point as attribute difference features between the first observation image and the second observation image.

[0057] In an optional implementation, when determining the target difference degree between the first observation image and the second observation image based on the attribute difference features between the first observation image and the second observation image, the detection module is configured to:

[0058] for each key point, determine an attribute difference weight of the key point based on the target ratio of the key point;

[0059] select a plurality of target key points from the plurality of key points based on the attribute difference weights of the key points, so that a difference evaluation value of a cube composed of the target key points is maximum, and the difference evaluation value is determined based on the attribute difference weights of the key points in the cube;

[0060] determine the target difference degree between the first observation image and the second observation image based on a ratio between a median of side lengths of the cube composed of the target key points and a first dimension of any pixel point in the first observation image or the second observation image in a world coordinate system.

[0061] In an optional implementation, when determining the target difference degree between the first observation image and the second observation image based on the geometric difference features and the attribute difference features between the first observation image and the second observation image, the detection module is configured to:

[0062] determine a first difference degree between the first observation image and the second observation image based on the geometric difference features;

[0063] determine a second difference degree between the first observation image and the second observation image based on the attribute difference features;

[0064] determine the target difference degree between the first observation image and the second observation image based on the first difference degree and the second difference degree.

[0065] In a third aspect, the optional implementation of the present disclosure further provides an electronic device, a processor and a memory, the memory stores machine readable instructions executable by the processor, and the processor is configured to execute the machine readable instructions stored in the memory, and when the machine readable instructions are executed by the processor, the machine readable instructions are executed to perform the steps of the first aspect or any possible implementation of the first aspect.

[0066] In a fourth aspect, the optional implementation of the present disclosure further provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and when the computer program is executed, the steps of the first aspect or any possible implementation of the first aspect are executed.

[0067] The method and device for detecting the level of detail model provided by the embodiments of the present disclosure obtain a standard model and a level of detail model of a target virtual object; pre-process the standard model and the level of detail model to obtain a first observation image of the standard model in an observation coordinate system matched with the level of detail model, and a second observation image of the level of detail model in the observation coordinate system; determine a target difference degree between the first observation image and the second observation image based on geometric difference features and / or attribute difference features between the first observation image and the second observation image; and determine that the level of detail model fails the quality detection in a case where the target difference degree is greater than or equal to a qualified threshold. The embodiments of the present disclosure pre-process the level of detail model and the standard model to obtain the first observation image and the second observation image in the observation coordinate system, and then determine the target difference degree based on the geometric difference features and / or the attribute difference features between the first observation image and the second observation image, so that the target difference degree is more accurate, and the accuracy of the quality detection is improved, thereby improving the quality of the level of detail model.

[0068] In order to make the above objectives, features and advantages of the present disclosure more obvious and easy to understand, the following preferred embodiments are specifically described below, and the accompanying drawings are described in detail as follows. BRIEF DESCRIPTION OF DRAWINGS

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following will briefly introduce the drawings needed to be used in the embodiments. The drawings herein are incorporated into the specification and form a part of the specification, which illustrate the embodiments consistent with the present disclosure, and are used to explain the technical solutions of the present disclosure together with the specification. It should be understood that the following drawings only show some embodiments of the present disclosure, and therefore should not be considered as a limitation to the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0070] Figure 1A flow chart of a level of detail model detection method provided by an embodiment of the present disclosure is shown.

[0071] Figure 2 A schematic diagram of a level of detail model detection device provided by an embodiment of the present disclosure is shown.

[0072] Figure 3 A schematic diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0073] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be described below in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. The components of the embodiments of the present disclosure described and shown herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure is not intended to limit the scope of the claimed present disclosure, but only represents selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present disclosure.

[0074] It is found through research that in a virtual three-dimensional scene, some virtual objects use level of detail (LOD) models. Compared with the standard model of a virtual object, an LOD model has fewer planes. When there is a certain distance between a camera and a virtual object, an LOD model corresponding to the distance range is usually used for picture preprocessing. The farther the distance, the fewer the visible planes of the virtual object, and thus the fewer the corresponding LOD model. Each LOD model corresponds to a distance range. When the distance between the camera and the virtual object exceeds the distance range of the current LOD model, LOD switching is performed. Therefore, as the distance changes, the LOD model is switched accordingly. Since the LOD model is relatively complex, deformation may occur when it is generated. Different LOD models may have large differences from the standard model, and picture jumps may occur when switching. If the jumps are too obvious, the visual effect of the virtual object will be greatly damaged.

[0075] Based on the above research, the present disclosure provides a level of detail model detection method. By preprocessing a level of detail model and a standard model, a first observation image and a second observation image in an observation coordinate system are obtained. Then, a target difference degree is determined based on geometric difference features and / or attribute difference features between the second observation image and the second observation image. The target difference degree is more accurate, and thus the accuracy of quality detection is improved, which facilitates improving the quality of the level of detail model.

[0076] The above-mentioned defects are the results of the inventors' practice and careful research, and thus the discovery process of the above-mentioned problems and the solutions proposed by the present disclosure to solve the above-mentioned problems should be the contributions of the inventors to the present disclosure.

[0077] It should be noted that similar reference numerals and letters represent similar items in the following drawings, and thus, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.

[0078] To facilitate the understanding of the present embodiment, first, a detailed hierarchical model detection method disclosed by the present embodiment is introduced in detail, and the execution subject of the detailed hierarchical model detection method provided by the present embodiment is generally a computer device with certain computing power. In some possible implementation manners, the detailed hierarchical model detection method can be realized by a processor calling computer readable instructions stored in a memory.

[0079] The detailed hierarchical model detection method provided by the present embodiment is described below taking a computer as an example.

[0080] Referring to FIG. 1, Figure 1 The method includes steps S101-S104, wherein:

[0081] S101, obtaining a standard model and a detailed hierarchical model of a target virtual object.

[0082] The target virtual object can be a three-dimensional virtual object, which can be deployed in a three-dimensional scene such as a game scene, a virtual reality scene, and an augmented reality scene. The target virtual object corresponds to a standard model and one or more detailed hierarchical models. The detailed hierarchical model can be obtained based on the standard model, and the number of planes of the detailed hierarchical model is lower than that of the standard model.

[0083] S102, preprocessing the standard model and the detailed hierarchical model to obtain a first observation image of the standard model in an observation coordinate system matched with the detailed hierarchical model, and a second observation image of the detailed hierarchical model in the observation coordinate system.

[0084] After obtaining the detailed hierarchical model and the standard model, the standard model and the detailed hierarchical model can be preprocessed in an observation coordinate system, and the preprocessing results can be observed in the observation coordinate system matched with the detailed hierarchical model to obtain a first observation image corresponding to the standard model and a second observation image corresponding to the detailed hierarchical model, respectively. The camera position for observation can be any position within the distance range corresponding to the detailed hierarchical model, and the observation image can be a three-dimensional image.

[0085] S103, determine a target difference degree between the first observation image and the second observation image based on the geometric difference feature and / or the attribute difference feature between the first observation image and the second observation image.

[0086] The geometric difference feature can be a difference feature in shape structure between the first observation image and the second observation image, and the attribute difference feature can be a difference feature in color and texture map between the first observation image and the second observation image.

[0087] If the first observation image and the second observation image are three-dimensional images, the geometric difference feature can include a first geometric difference feature of each first plane vertex in the first observation image and a second geometric difference feature of each second plane vertex in the second observation image.

[0088] In the observation image, the surface of each model can be referred to as a plane, and the vertex of each surface can be referred to as a plane vertex. The first geometric difference feature corresponds to the plane vertex in the first observation image, and any plane vertex in the first observation image can be referred to as a first plane vertex. Similarly, any plane vertex in the second observation image can be referred to as a second plane vertex.

[0089] To obtain the geometric difference feature between the first observation image and the second observation image, the positions of the first observation image and the second observation image in the observation coordinate system can be adjusted so that the center points of the first observation image and the second observation image coincide, that is, the coordinates of the center points of the standard model and the level-of-detail model are the same in the observation coordinate system.

[0090] Then, the first distance of each first plane vertex to each second plane in the second observation image can be determined, and the maximum first distance of each first plane vertex is taken as the first geometric difference feature of the first plane vertex, that is, the distance between the first plane vertex and the closest second plane is taken as the first geometric difference feature of the first plane vertex. Since the level-of-detail model is obtained by subtracting the surface of the standard model, it is difficult to determine the correspondence between the vertices and the surfaces of the level-of-detail model and the standard model, and it is also difficult to determine the relative objects for comparison between the two models and the difference between the two models. Therefore, the distance between the closest point and the surface can be taken as the geometric difference feature, which can effectively reflect the difference between the two models.

[0091] Similarly, the second distance of each second plane vertex in the second observation image to each first plane in the first observation image can be determined, and the minimum second distance of each second plane vertex in the second observation image is taken as the second geometric difference feature of the second plane vertex, which can play the same role.

[0092] In the case of the geometric difference feature, the target geometric difference feature can be determined from the maximum geometric difference feature in the first set of difference features and the second geometric difference features, and then the first size corresponding to the coordinate range of any pixel point in the first observation image or the second observation image in the world coordinate system is determined.

[0093] For example, the second size of any pixel point in the first observation image or the second observation image, the picture height of the first observation image or the second observation image, the shooting angle parameter of the camera when the standard model or the detail level model is preprocessed, and the third distance between the camera and the standard model or the detail level model can be obtained. Since the first observation image and the second observation image are both observed in the observation coordinate system, their second sizes, picture heights, shooting angle parameters and third distances are consistent. Then, the ratio of the camera angle to π / 360 can be calculated, and the tangent value of the calculated ratio is multiplied by 2, and then multiplied by the ratio between the third distance and the picture height, and the obtained value is taken as the first size. The shooting angle parameter can be the field of view angle, and the field of view angle can be the angle between the upper and lower two shooting boundary planes of the camera.

[0094] After obtaining the first size, the ratio between the target geometric difference feature and the first size can be calculated, and the obtained ratio is taken as the target difference degree.

[0095] Correspondingly, the attribute difference feature between the first observation image and the second observation image can be obtained by the following steps:

[0096] The first observation image and the second observation image are converted into voxels to obtain a first mesh model corresponding to the first observation image and a second mesh model corresponding to the second observation image;

[0097] For each key point, a first plane in which the key point is located in the first mesh model is determined, and a second plane in which the key point is located in the second mesh model is determined;

[0098] Based on the attribute information of the first mesh matched with each first plane vertex of the first plane, the attribute information of the key point in the first observation image is determined;

[0099] Based on the attribute information of the second mesh matched with each second plane vertex of the second plane, the attribute information of the key point in the second observation image is determined;

[0100] determine a target ratio between the attribute information of the key point in the first observation image and the attribute information of the key point in the second observation image, and take the target ratio of each key point as the attribute difference feature between the first observation image and the second observation image.

[0101] The first observation image and the second observation image can include a three-dimensional model image, the three-dimensional model image can be converted into voxels to form a plurality of voxel points, and a first mesh model and a second mesh model composed of key points are obtained. Then, a plurality of key points can be determined from the coordinate system corresponding to the mesh model, and the key points can be preset fixed points. After determining the key points, the planes where the first mesh model and the second mesh model where the key points are located can be determined. The attribute information of the first mesh matched at each vertex of the first plane of the first mesh model based on the key points is determined as the attribute information of the key point in the first observation image.

[0102] The attribute information of the first mesh includes the coordinates of the texture image corresponding to the vertex. Through the attribute information of each first mesh, the attribute information of the center point of the corresponding plane of each first mesh can be determined, and the attribute information is taken as the attribute information of the key point in the first observation image.

[0103] Correspondingly, the attribute information of each key point in the second observation image can be determined in a similar manner.

[0104] After obtaining the attribute information of the key points in the first observation image and the second observation image respectively, the target ratio between the attribute information of the key points in the first observation image and the attribute information of the key points in the second observation image can be determined, and the target ratio is taken as the attribute information of the key point.

[0105] After determining the attribute information, the target ratio between the attribute information of the key points in the first observation image and the attribute information of the key points in the second observation image is determined, and the determined target ratio is taken as the attribute difference value between the first observation image and the second observation image. The attribute difference value includes the target ratio of each key point.

[0106] Then, for each key point, a plurality of adjacent target key points can be selected from a plurality of key points based on the corresponding attribute difference weight of the key point to form a cube, so that the difference evaluation value of the cube is maximum. The difference evaluation value can be determined based on the attribute difference weight of each key point in the cube, and the target weight value can be a weight value greater than a preset value. In this way, the cube is the part with the largest attribute difference between the two models.

[0107] Exemplarily, the difference evaluation value can be calculated using a reward function, and by optimizing the reward function, the cube can contain as many key points equal to the target weight value as possible and as few key points not equal to the target weight value as possible.

[0108] Then, a ratio between the median of the side length of the cube and the first size can be determined as the target difference degree between the first observation image and the second observation image.

[0109] If the target difference degree is determined by using both the geometric difference feature and the attribute difference feature, the first difference degree corresponding to the geometric difference feature and the second difference degree corresponding to the attribute difference feature can be determined respectively by using the above steps, and then the target difference degree can be determined based on the first difference degree and the second difference degree.

[0110] Exemplarily, the target difference degree can be determined by weighted sum of the first difference degree and the second difference degree, or the target difference degree can be the difference degree with the maximum median of the first difference degree and the second difference degree.

[0111] In S104, if the target difference degree is greater than or equal to the qualified threshold, it is determined that the level-of-detail model fails the quality detection.

[0112] If the target difference degree is greater than or equal to the qualified threshold, it indicates that the difference between the level-of-detail model and the standard model is too large, and the visual effect is poor. Therefore, it can be determined that the level-of-detail model fails the quality detection, so that the user can modify the level-of-detail model to optimize the visual effect.

[0113] The level-of-detail model detection method provided by the embodiments of the present disclosure includes the following steps: obtaining a standard model and a level-of-detail model of a target virtual object; preprocessing the standard model and the level-of-detail model to obtain a first observation image of the standard model in an observation coordinate system matched with the level-of-detail model and a second observation image of the level-of-detail model in the observation coordinate system; determining a target difference degree between the first observation image and the second observation image based on geometric difference features and / or attribute difference features between the first observation image and the second observation image; and determining that the level-of-detail model fails the quality detection if the target difference degree is greater than or equal to a qualified threshold. The embodiments of the present disclosure preprocess the level-of-detail model and the standard model to obtain the first observation image and the second observation image in the observation coordinate system, and then determine the target difference degree based on the geometric difference features and / or the attribute difference features between the first observation image and the second observation image, so that the target difference degree is more accurate, and the accuracy of the quality detection is improved, thereby improving the quality of the level-of-detail model.

[0114] Those skilled in the art can understand that the sequence of writing each step in the above method of the specific embodiment does not mean a strict execution sequence and does not constitute any limitation on the implementation process, and the specific execution sequence of each step should be determined by its function and possible internal logic.

[0115] Based on the same inventive concept, the disclosure embodiments also provide a detail level model detection device corresponding to the detail level model detection method. Since the principle of solving problems in the device of the disclosure embodiments is similar to the above-mentioned detail level model detection method of the disclosure embodiments, the implementation of the device can be referred to the implementation of the method, and the repeated parts will not be described here.

[0116] Referring to Figure 2 Fig. 2 shows a schematic diagram of a detail level model detection device provided by the disclosure embodiments, which includes:

[0117] The acquisition module 210 is configured to acquire a standard model and a detail level model of a target virtual object.

[0118] The preprocessing module 220 is configured to pre-process the standard model and the detail level model to obtain a first observation image of the standard model in an observation coordinate system matched with the detail level model, and a second observation image of the detail level model in the observation coordinate system.

[0119] The detection module 230 is configured to determine a target difference degree between the first observation image and the second observation image based on geometric difference features and / or attribute difference features between the first observation image and the second observation image.

[0120] The judgment module 240 is configured to determine that the detail level model fails the quality detection if the target difference degree is greater than or equal to a qualified threshold.

[0121] In an optional implementation, the geometric difference features between the first observation image and the second observation image include first geometric difference features of each first plane vertex in the first observation image, and second geometric difference features of each second plane vertex in the second observation image.

[0122] When acquiring the geometric difference features between the first observation image and the second observation image, the acquisition module 210 is configured to:

[0123] Adjust the positions of the first observation image and the second observation image in the observation coordinate system so that the center points of the first observation image and the second observation image coincide.

[0124] determining a first distance from each first plane vertex in the first observation image to a respective second plane in the second observation image; for each first plane vertex in the first observation image, taking the smallest first distance as a first geometric difference feature of the first plane vertex; and,

[0125] determining a second distance from each second plane vertex in the second observation image to a respective first plane in the first observation image; for each second plane vertex in the second observation image, taking the smallest second distance as a second geometric difference feature of the second plane vertex.

[0126] In an optional implementation, when determining the target difference degree between the first observation image and the second observation image based on the geometric difference features between the first observation image and the second observation image, the detection module 230 is configured to:

[0127] determining a target geometric difference feature as the maximum of the first geometric difference feature and the second geometric difference feature;

[0128] determining the target difference degree between the first observation image and the second observation image based on a first size corresponding to a coordinate range of any pixel in the first observation image or the second observation image in a world coordinate system, and the target geometric difference feature.

[0129] In an optional implementation, when determining the first size corresponding to the coordinate range of any pixel in the first observation image or the second observation image in the world coordinate system, the detection module 230 is configured to:

[0130] obtaining a second size of any pixel in the first observation image or the second observation image, a picture height of the first observation image or the second observation image, a shooting angle parameter of a camera when pre-processing the standard model or the detail level model, and a third distance between the camera and the standard model or the detail level model;

[0131] determining the first size based on a ratio between the second size, the shooting angle parameter, and the third distance and the picture height.

[0132] In an optional implementation, when obtaining the attribute difference feature between the first observation image and the second observation image, the detection module 230 is configured to:

[0133] convert the first observation image and the second observation image into voxels to obtain a first mesh model corresponding to the first observation image and a second mesh model corresponding to the second observation image;

[0134] For each key point, determine a first plane in which the key point is located in the first mesh model, and a second plane in which the key point is located in the second mesh model;

[0135] Based on attribute information of the first mesh matched with each first plane vertex of the first plane, determine attribute information of the key point in the first observation image;

[0136] Based on attribute information of the second mesh matched with each second plane vertex of the second plane, determine attribute information of the key point in the second observation image;

[0137] Determine a target ratio between the attribute information of the key point in the first observation image and the attribute information of the key point in the second observation image, and take the target ratio of each key point as an attribute difference feature between the first observation image and the second observation image.

[0138] In an optional implementation, when the detection module 230 determines the target difference degree between the first observation image and the second observation image based on the attribute difference feature between the first observation image and the second observation image, the detection module 230 is configured to:

[0139] For each key point, determine an attribute difference weight of the key point based on the target ratio of the key point;

[0140] Based on the attribute difference weights of each key point, filter a plurality of target key points from the plurality of key points, so that a difference evaluation value of a cube composed of the target key points is maximum, and the difference evaluation value is determined based on the attribute difference weights of each key point in the cube;

[0141] Based on a ratio between a median of side lengths of the cube composed of the target key points and a first size of any pixel point in the world coordinate system in the first observation image or the second observation image, determine the target difference degree between the first observation image and the second observation image.

[0142] In an optional implementation, when the detection module determines the target difference degree between the first observation image and the second observation image based on the geometric difference feature and the attribute difference feature between the first observation image and the second observation image, the detection module is configured to:

[0143] Based on the geometric difference feature, determine a first difference degree between the first observation image and the second observation image;

[0144] Based on the attribute difference feature, determine a second difference degree between the first observation image and the second observation image;

[0145] determine a target difference degree between the first observation image and the second observation image based on the first difference degree and the second difference degree.

[0146] The description of the processing flow of each module in the device and the interaction flow between the modules can refer to the related description in the above method embodiments, and will not be described in detail here.

[0147] The embodiments of the present disclosure also provide an electronic device, such as Figure 3 As shown in the figure, the electronic device structure schematic diagram provided by the embodiments of the present disclosure includes:

[0148] a processor 31 and a memory 32; the memory 32 stores machine readable instructions executable by the processor 31, and the processor 31 is configured to execute the machine readable instructions stored in the memory 32, and when the machine readable instructions are executed by the processor 31, the processor 31 executes the following steps:

[0149] obtain a standard model and a detail level model of a target virtual object;

[0150] preprocess the standard model and the detail level model to obtain a first observation image of the standard model in an observation coordinate system matched with the detail level model, and a second observation image of the detail level model in the observation coordinate system;

[0151] determine a target difference degree between the first observation image and the second observation image based on geometric difference features and / or attribute difference features between the first observation image and the second observation image;

[0152] if the target difference degree is greater than or equal to a qualified threshold, determine that the detail level model fails the quality detection.

[0153] In an optional implementation, the geometric difference features between the first observation image and the second observation image in the instructions executed by the processor 31 include first geometric difference features of each first plane vertex in the first observation image, and second geometric difference features of each second plane vertex in the second observation image;

[0154] When the processor 31 executes the geometric difference features between the first observation image and the second observation image, it is configured to execute:

[0155] adjust the positions of the first observation image and the second observation image in the observation coordinate system so that the center points of the first observation image and the second observation image coincide;

[0156] determining a first distance from each first plane vertex in the first observation image to a respective second plane in the second observation image; for each first plane vertex in the first observation image, taking the smallest first distance as a first geometric difference feature of the first plane vertex; and,

[0157] determining a second distance from each second plane vertex in the second observation image to a respective first plane in the first observation image; for each second plane vertex in the second observation image, taking the smallest second distance as a second geometric difference feature of the second plane vertex.

[0158] In an optional implementation, the instructions executed by the processor 31 include determining a target difference degree between the first observation image and the second observation image based on the geometric difference features between the first observation image and the second observation image, including:

[0159] determining a target geometric difference feature as the maximum of the first geometric difference feature and the second geometric difference feature;

[0160] determining a target difference degree between the first observation image and the second observation image based on a first size corresponding to a coordinate range of any pixel in the first observation image or the second observation image in a world coordinate system, and the target geometric difference feature.

[0161] In an optional implementation, the instructions executed by the processor 31 include determining the first size corresponding to the coordinate range of any pixel in the first observation image or the second observation image in the world coordinate system, including:

[0162] obtaining a second size of any pixel in the first observation image or the second observation image, a picture height of the first observation image or the second observation image, a shooting angle parameter of a camera when preprocessing the standard model or the detail level model, and a third distance between the camera and the standard model or the detail level model;

[0163] determining the first size based on a ratio between the second size, the shooting angle parameter, and the third distance and the picture height.

[0164] In an optional implementation, the instructions executed by the processor 31 include obtaining an attribute difference feature between the first observation image and the second observation image, including:

[0165] converting the first observation image and the second observation image into voxels to obtain a first mesh model corresponding to the first observation image and a second mesh model corresponding to the second observation image;

[0166] For each key point, determine a first plane in which the key point is located in the first mesh model, and a second plane in which the key point is located in the second mesh model;

[0167] Based on attribute information of the first mesh matched with each first plane vertex of the first plane, determine attribute information of the key point in the first observation image;

[0168] Based on attribute information of the second mesh matched with each second plane vertex of the second plane, determine attribute information of the key point in the second observation image;

[0169] Determine a target ratio between the attribute information of the key point in the first observation image and the attribute information of the key point in the second observation image, and take the target ratio of each key point as an attribute difference feature between the first observation image and the second observation image.

[0170] In an optional implementation, the instructions executed by the processor 31 include determining a target difference degree between the first observation image and the second observation image based on the attribute difference feature between the first observation image and the second observation image, including:

[0171] For each key point, determine an attribute difference weight of the key point based on the target ratio of the key point;

[0172] Based on the attribute difference weights of each key point, filter a plurality of target key points from the plurality of key points, so that a difference evaluation value of a cube composed of the target key points is maximum, and the difference evaluation value is determined based on the attribute difference weights of each key point in the cube;

[0173] Based on a ratio between a median of side lengths of the cube composed of the target key points and a first size of any pixel point in the first observation image or the second observation image in a world coordinate system, determine the target difference degree between the first observation image and the second observation image.

[0174] In an optional implementation, the instructions executed by the processor 31 include determining a target difference degree between the first observation image and the second observation image based on the geometric difference feature and the attribute difference feature between the first observation image and the second observation image, including:

[0175] Based on the geometric difference feature, determine a first difference degree between the first observation image and the second observation image;

[0176] determine a second difference degree between the first observation image and the second observation image based on the attribute difference feature;

[0177] determine a target difference degree between the first observation image and the second observation image based on the first difference degree and the second difference degree.

[0178] The memory 32 includes an internal memory 321 and an external memory 322. The internal memory 321 is also referred to as an internal storage, and is used to temporarily store operation data in the processor 31 and exchange data with the external memory 322 such as a hard disk. The processor 31 exchanges data with the external memory 322 through the internal memory 321.

[0179] The specific execution process of the instructions can refer to the steps of the detail-level model detection method described in the embodiments of the present disclosure, which will not be described here.

[0180] The embodiments of the present disclosure also provide a computer readable storage medium, which stores a computer program. When the computer program is run by a processor, the steps of the detail-level model detection method described in the above method embodiments are executed. The storage medium can be a volatile or non-volatile computer readable storage medium.

[0181] The computer program product of the detail-level model detection method provided by the embodiments of the present disclosure includes a computer readable storage medium storing program codes. The instructions included in the program codes can be used to execute the steps of the detail-level model detection method described in the above method embodiments. For details, refer to the above method embodiments, which will not be described here.

[0182] The embodiments of the present disclosure also provide a computer program, which is executed by a processor to implement any one of the methods of the preceding embodiments. The computer program product can be specifically implemented by hardware, software or a combination thereof. In one optional embodiment, the computer program product is specifically embodied as a computer storage medium. In another optional embodiment, the computer program product is specifically embodied as a software product, such as a software development kit (Software Development Kit, SDK) and the like.

[0183] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here. In several embodiments provided in the present disclosure, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and another division can be made in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some communication interfaces, devices or units, and can be electrical, mechanical or other forms.

[0184] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0185] In addition, each functional unit in each embodiment of the present disclosure can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0186] If the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present disclosure essentially or the part of the prior art or the part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present disclosure. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and various program code storage media.

[0187] Finally, it should be noted that the above-described embodiments are merely specific embodiments of the present disclosure, used to illustrate the technical solutions of the present disclosure, and are not intended to limit the present disclosure. The protection scope of the present disclosure is not limited thereto. Although the present disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can make modifications or easy changes to the technical solutions described in the foregoing embodiments, or easily think of changes or equivalent replacements for some of the technical features; and these modifications, changes or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A level of detail model detection method characterized by, The method comprises: obtaining a standard model and a detailed level model of a target virtual object, wherein the number of planes of the detailed level model is lower than that of the standard model; preprocessing the standard model and the detailed level model to obtain a first observation image of the standard model in an observation coordinate system matched with the detailed level model, and a second observation image of the detailed level model in the observation coordinate system, wherein the camera position for observation is a position within the distance range corresponding to the detailed level model; determining a target difference degree between the first observation image and the second observation image based on geometric difference features and / or attribute difference features between the first observation image and the second observation image; in a case where the target difference degree is greater than or equal to a qualified threshold, determining that the detailed level model fails the quality detection, wherein the geometric difference features between the first observation image and the second observation image include first geometric difference features of each first plane vertex in the first observation image, and second geometric difference features of each second plane vertex in the second observation image, wherein for each first plane vertex in the first observation image, the smallest first distance from each first plane vertex in the first observation image to each second plane in the second observation image is taken as the first geometric difference feature of the first plane vertex, wherein the target difference degree between the first observation image and the second observation image is determined based on the geometric difference features between the first observation image and the second observation image, comprising: determining the maximum geometric difference feature among the first geometric difference features and the second geometric difference features as a target geometric difference feature; determining the target difference degree between the first observation image and the second observation image based on a first size corresponding to the coordinate range of any pixel point in the first observation image or the second observation image in the world coordinate system, and the target geometric difference feature, wherein the first size corresponding to the coordinate range of any pixel point in the first observation image or the second observation image in the world coordinate system is determined by the following steps: obtaining a second size of any pixel point in the first observation image or the second observation image, a picture height of the first observation image or the second observation image, a shooting angle parameter of a camera when preprocessing the standard model or the detailed level model, and a third distance between the camera and the standard model or the detailed level model; determining the first size based on the ratio between the second size, the shooting angle parameter, and the third distance and the picture height, wherein the second size, the picture height, the shooting angle parameter, and the third distance of the first observation image and the second observation image are consistent.

2. The method of claim 1, wherein, The geometric difference features between the first observation image and the second observation image are obtained by the following steps: adjusting positions of the first observation image and the second observation image in the observation coordinate system so that center points of the first observation image and the second observation image coincide; and, determining a second distance from each second plane vertex in the second observation image to a respective first plane in the first observation image; for each second plane vertex in the second observation image, taking the smallest second distance as a second geometric difference feature of the second plane vertex.

3. The method of claim 1, wherein, obtaining attribute difference features between the first observation image and the second observation image by the following steps: converting the first observation image and the second observation image into voxels to obtain a first mesh model corresponding to the first observation image and a second mesh model corresponding to the second observation image; for each key point, determining a first plane in which the key point is located in the first mesh model, and a second plane in which the key point is located in the second mesh model; determining attribute information of the key point in the first observation image based on attribute information of a first mesh matched with each first plane vertex of the first plane; determining attribute information of the key point in the second observation image based on attribute information of a second mesh matched with each second plane vertex of the second plane; determining a target ratio between the attribute information of the key point in the first observation image and the attribute information of the key point in the second observation image, and taking the target ratio of each key point as an attribute difference feature between the first observation image and the second observation image.

4. The method of claim 3, wherein, determining a target difference degree between the first observation image and the second observation image based on the attribute difference features between the first observation image and the second observation image, comprising: for each key point, determining an attribute difference weight of the key point based on the target ratio of the key point; based on the attribute difference weights of each key point, selecting a plurality of adjacent target key points from the plurality of key points so that a difference evaluation value of a cube composed of the target key points is maximum, the difference evaluation value being determined based on the attribute difference weights of each key point in the cube; determining the target difference degree between the first observation image and the second observation image based on a ratio between a median of side lengths of the cube composed of the target key points and a first size of any pixel point in the first observation image or the second observation image in a world coordinate system.

5. The method of claim 1, wherein, determining a target difference degree between the first observation image and the second observation image based on the geometric difference features and the attribute difference features between the first observation image and the second observation image, comprising: determining a first difference degree between the first observation image and the second observation image based on the geometric difference features; determining a second difference degree between the first observation image and the second observation image based on the attribute difference features; determining the target difference degree between the first observation image and the second observation image based on the first difference degree and the second difference degree.

6. A detail level model detection apparatus characterized by comprising: comprising: An acquisition module is configured to acquire a standard model and a detail level model of a target virtual object, wherein a number of planes of the detail level model is less than a number of planes of the standard model; A preprocessing module is configured to preprocess the standard model and the detail level model to obtain a first observation image of the standard model in an observation coordinate system matched with the detail level model and a second observation image of the detail level model in the observation coordinate system, wherein a camera position for observation is a position within a distance range corresponding to the detail level model; A detection module is configured to determine a target difference degree between the first observation image and the second observation image based on geometric difference features and / or attribute difference features between the first observation image and the second observation image; A judgment module is configured to determine that the detail level model fails to pass quality detection if the target difference degree is greater than or equal to a qualified threshold. The geometric difference features between the first observation image and the second observation image include first geometric difference features of each first plane vertex in the first observation image and second geometric difference features of each second plane vertex in the second observation image, wherein for each first plane vertex in the first observation image, a smallest first distance from each first plane vertex in the first observation image to each second plane in the second observation image is taken as the first geometric difference feature of the first plane vertex. The detection module is further configured to: determine a target geometric difference feature as a maximum geometric difference feature from the first geometric difference features and the second geometric difference features; and determine the target difference degree between the first observation image and the second observation image based on a first size corresponding to a coordinate range of any pixel in the first observation image or the second observation image in a world coordinate system and the target geometric difference feature. The first size corresponding to the coordinate range of any pixel in the first observation image or the second observation image in the world coordinate system is determined by: acquiring a second size of any pixel in the first observation image or the second observation image, a picture height of the first observation image or the second observation image, a shooting angle parameter of a camera when preprocessing the standard model or the detail level model, and a third distance between the camera and the standard model or the detail level model; and determining the first size based on a ratio between the second size, the shooting angle parameter, and the third distance and the picture height. The second size, the picture height, the shooting angle parameter, and the third distance of the first observation image and the second observation image are consistent.

7. An electronic device, comprising: The method comprises: A processor, a memory, the memory storing machine readable instructions executable by the processor, the processor being configured to execute the machine readable instructions stored in the memory, the machine readable instructions, when executed by the processor, causing the processor to perform the steps of the detail level model detection method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, A computer program is stored in the computer readable storage medium, and when the computer program is run by a computer device, the computer device performs the steps of the detail level model method according to any one of claims 1 to 5.

Images