AR testing method, device and computer-readable storage medium
By scanning the identification images displayed by the second test terminal in different test scenarios, the feature point information is obtained, and the WebGL technology and AR software development toolkit are used to generate superimposed images to evaluate the functions of AR applications, which solves the problem of low accuracy of existing AR testing methods. The problem has achieved more rigorous and reliable test results.
Patent Information
- Application Number
- CN202111046731.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-09-07
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2041-09-07
AI Technical Summary
The existing AR testing methods mainly rely on manual subjective evaluation, resulting in low accuracy of test results and lack of rigor and operability.
Feature point information is obtained by scanning the identification images displayed by the second test terminal in different test scenarios, using 3D graphics built based on the 3D drawing protocol WebGL, combining the AR software development toolkit to be detected to identify feature point information, and generating different test scenarios and evaluate the functionality of the AR application based on the test parameters.
It improves the accuracy and rigor of AR testing, and evaluates the functions of AR applications through test parameters obtained through multi-scene testing, providing more objective and reliable test results.
Smart Images

Figure CN113887566B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of testing technology, and in particular to an AR testing method, device, and computer-readable storage medium. Background Art
[0002] At present, in order to facilitate AR (Augmented Reality) developers, many companies have launched software development kits (SDKs) specifically for the development of augmented reality technology, such as Qualcomm vuforia SDK, AR Toolkit, and the more popular EasyAR SDK and HiAR SDK in China.
[0003] Although there are many mainstream AR software development kits on the market, there is a large gap in the testing technology and methods of AR software development kits. Existing AR testing is mainly implemented manually, and most of them use subjective evaluation to judge the quality of the engine. In this way, the test results are subjective, so the accuracy of the test results is low. Summary of the invention
[0004] The embodiments of the present application aim to solve the problem of low accuracy of existing AR test results by providing an AR test method, device and computer-readable storage medium.
[0005] To achieve the above objectives, the present application provides an AR testing method, which includes:
[0006] Scanning the identification pictures displayed by the second test terminal in different test scenarios to obtain feature point information, wherein the display parameters of the identification pictures in the different test scenarios are different, and the identification pictures are 3D graphics constructed based on the 3D drawing protocol WebGL and stored in the target library to be identified of the second test terminal;
[0007] Acquiring a model to be loaded according to the feature point information, wherein recognition of the feature point information is realized by an AR software development kit to be detected;
[0008] Superimposing the model to be loaded with the feature point information to generate superimposed images under different test scenarios;
[0009] Acquire test parameters of each of the superimposed images, and evaluate the AR function of the AR application according to the test parameters.
[0010] Optionally, the step of evaluating the AR function of the AR application according to the test parameters includes:
[0011] Get the weight values of different test scenarios;
[0012] Acquire test scores of the AR function in different test scenarios according to the test parameters;
[0013] The test scores are weighted and summed according to the weight values to obtain a target score, and the AR function of the AR application is evaluated according to the target score.
[0014] Optionally, the step of acquiring the test scores of the AR function in different test scenarios according to the test parameters includes:
[0015] Obtaining test scoring rules and test parameters corresponding to the type of each test scenario;
[0016] Under each type of the test scenario, the test parameters of the test scenario are scored for AR function testing according to the test scoring rules corresponding to the test scenario to obtain test scores of the AR function of the AR application under different test scenarios.
[0017] Optionally, the step of evaluating the AR function of the AR application according to the target score includes:
[0018] Obtaining the score level corresponding to the target score;
[0019] The AR function of the AR application is evaluated according to the score level.
[0020] Optionally, before the step of scanning the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, the step includes:
[0021] Get pre-built 3D models;
[0022] Identify the logo image according to the AR software development kit to be detected;
[0023] The three-dimensional model is bound to the identified logo image, and an AR application package is exported according to the binding information.
[0024] Optionally, the step of superimposing the model to be loaded with the feature point information to generate superimposed images under different test scenarios includes:
[0025] Rendering the model to be loaded into a virtual image;
[0026] The virtual image and the feature point information are superimposed to generate superimposed images under different test scenarios.
[0027] Optionally, the step of acquiring the model to be loaded according to the feature point information includes:
[0028] Matching the feature point information with pre-stored feature point information to obtain a matching result;
[0029] When the matching result is that the feature point information matches the pre-stored feature point information, obtaining a model to be loaded associated with the pre-stored feature point information;
[0030] Wherein, the feature point information is associated with the model to be loaded in advance.
[0031] In addition, to achieve the above-mentioned purpose, the present application also provides an AR testing device on the other hand, which includes a memory, a processor, and an AR testing program stored in the memory and running on the processor, and when the processor executes the AR testing program, the steps of the AR testing method described above are implemented.
[0032] In addition, to achieve the above-mentioned purpose, the present application further provides an AR testing device, which includes a scanning module, an acquisition module, a superposition module and an evaluation module, wherein:
[0033] The scanning module is used to scan the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, the display parameters of the identification picture in different test scenarios are different, the identification picture is a 3D graphic constructed based on the 3D drawing protocol WebGL, and is stored in the target library to be identified of the second test terminal;
[0034] The acquisition module is used to acquire the model to be loaded according to the feature point information, and the recognition of the feature point information is realized by the AR software development kit to be detected;
[0035] The superposition module is used to superimpose the model to be loaded with the feature point information to generate superimposed images under different test scenarios;
[0036] The evaluation module is used to obtain test parameters of each of the superimposed images and evaluate the AR function of the AR application according to the test parameters.
[0037] In addition, to achieve the above-mentioned purpose, the present application also provides a computer-readable storage medium on the other hand, on which an AR test program is stored, and when the AR test program is executed by a processor, the steps of the AR test method described above are implemented.
[0038] The present application proposes an AR testing method, which obtains feature point information by scanning the identification picture displayed by the second test terminal in different test scenarios. The display parameters of the identification picture in different test scenarios are different. The identification picture is a 3D graphic built based on the 3D drawing protocol WebGL and stored in the target library to be identified of the second test terminal; the model to be loaded is obtained according to the feature point information, and the identification of the feature point information is realized by the AR software development kit to be detected; the model to be loaded is superimposed with the feature point information to generate superimposed images in different test scenarios; the test parameters of each superimposed image are obtained, and the AR function of the AR application is evaluated according to the test parameters. The present application performs multi-scenario testing on the AR application, obtains the test parameters of the superimposed images in different test scenarios, and evaluates the AR function of the AR application based on the test parameters, thereby improving the accuracy of the AR test. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] Figure 1 A schematic diagram of the terminal structure of the hardware operating environment involved in the embodiment of the present application;
[0040] Figure 2 This is a flow chart of the first embodiment of the AR testing method of the present application;
[0041] Figure 3 A schematic diagram of a process of evaluating the AR function of an AR application according to the test parameters in the AR test method of the present application;
[0042] Figure 4 This is a flow chart before the step of scanning the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information in the AR test method of the present application;
[0043] Figure 5 This is a schematic diagram of the operation flow of the AR testing method for this application;
[0044] Figure 6 This is a module diagram of the AR testing method of this application.
[0045] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0046] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0047] The main solution of the embodiment of the present application is: scanning the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, the display parameters of the identification picture in different test scenarios are different, the identification picture is a 3D graphic built based on the 3D drawing protocol WebGL, and is stored in the target library to be identified of the second test terminal; obtaining the model to be loaded according to the feature point information, and the identification of the feature point information is realized by the AR software development kit to be detected; superimposing the model to be loaded with the feature point information to generate superimposed images in different test scenarios; obtaining the test parameters of each of the superimposed images, and evaluating the AR function of the AR application according to the test parameters.
[0048] Since the existing AR tests are mainly implemented manually, most of them use subjective evaluation to judge the quality of the engine. In this way, the test results are subjective, so the accuracy of the test results is low. However, the present application obtains feature point information by scanning the identification pictures displayed by the second test terminal in different test scenarios. The display parameters of the identification pictures in different test scenarios are different. The identification pictures are 3D graphics built based on the 3D drawing protocol WebGL and stored in the target library to be identified of the second test terminal; the model to be loaded is obtained according to the feature point information, and the recognition of the feature point information is realized by the AR software development kit to be detected; the model to be loaded is superimposed with the feature point information to generate superimposed images in different test scenarios; the test parameters of each superimposed image are obtained, and the AR function of the AR application is evaluated according to the test parameters. The present application performs multi-scenario testing on the AR application, obtains the test parameters of the superimposed images in different test scenarios, and evaluates the AR function of the AR application based on the test parameters, thereby improving the accuracy of the AR test.
[0049] like Figure 1 As shown, Figure 1 This is a schematic diagram of the terminal structure of the hardware operating environment involved in the embodiment of the present application.
[0050] like Figure 1As shown, the terminal may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.
[0051] Those skilled in the art will understand that Figure 1 The terminal structure shown in the figure does not constitute a limitation on the terminal device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.
[0052] like Figure 1 As shown, the memory 1005 as a computer-readable storage medium may include an AR test program.
[0053] exist Figure 1 In the terminal shown, the network interface 1004 is mainly used for data communication with the background server; the user interface 1003 is mainly used for data communication with the client (user end); the processor 1001 can be used to call the AR test program in the memory 1005 and perform the following operations:
[0054] Scanning the identification pictures displayed by the second test terminal in different test scenarios to obtain feature point information, wherein the display parameters of the identification pictures in the different test scenarios are different, and the identification pictures are 3D graphics constructed based on the 3D drawing protocol WebGL and stored in the target library to be identified of the second test terminal;
[0055] Acquiring a model to be loaded according to the feature point information, wherein recognition of the feature point information is realized by an AR software development kit to be detected;
[0056] Superimposing the model to be loaded with the feature point information to generate superimposed images under different test scenarios;
[0057] Acquire test parameters of each of the superimposed images, and evaluate the AR function of the AR application according to the test parameters.
[0058] refer to Figure 2 , Figure 2This is a flowchart of the first embodiment of the AR testing method of the present application.
[0059] The embodiment of the present application provides an AR testing method. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order different from that shown here.
[0060] The AR testing method of this embodiment includes the following steps:
[0061] Step S10, scanning the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, wherein the display parameters of the identification picture in different test scenarios are different, and the identification picture is a 3D graphic constructed based on the 3D drawing protocol WebGL, and is stored in the target library to be identified of the second test terminal;
[0062] It should be noted that the first test terminal and the second test terminal of the present application can be a mobile phone, a tablet, a computer, etc., wherein an AR application is running in the first test terminal, and the AR application is developed according to the AR software development kit to be tested, that is, AR SDK (Software Development Kit, referred to as SDK, software development kit). The present application tests the AR application developed according to the AR software development kit to be tested to evaluate the overall advantages and disadvantages of the AR software development kit to be tested.
[0063] Due to the existing various scenarios of simulating interference with the marker (mainly picture information) manually or with the help of external devices, it lacks rigor and operability. Based on this problem, the present application is a logo recognition picture library module constructed by WebGL technology (3D drawing protocol), which is used to store and place the target picture library to be identified, and simulates various interference scenarios (i.e. test scenarios) for the logo picture based on the logo recognition picture library module. In one embodiment, WebGL technology is used to build a Web3D application. In order to realize the operation of the logo picture, the picture imported by the picture library is not a picture in the traditional jpg or png format, but a cube with a very small thickness, and the visual effect is the same as the picture. In this way, the mathematical functions in the open source framework Three.js of WebGL can be used to complete various operations on the "picture", for example, the PointLight attribute is used to control the intensity of the light, and THREE.Vector is used to complete the change of angle tilt, etc. The classified stored logo pictures are obtained from the logo recognition picture library module, and at the same time, the display parameters of the logo pictures under different test scenarios are obtained. The logo pictures under different test scenarios can be obtained by adjusting the logo pictures according to the display parameters. Among them, the test scene can be a test of lighting environment change, left and right movement tracking test, angle tilt test, multi-target recognition test, and identification object near and far view test; the display parameters of the logo picture can be angle, brightness, near and far view, etc. For example, according to the display parameters, the logo picture can be added with a light source, the angle can be adjusted, and the near and far distance can be adjusted, that is, interference factors can be added to the logo picture, so that the logo picture has abnormal lighting, large tilt angle, near and far distance, etc. Abnormal. In this way, the corresponding functions of the AR application in the test scene of abnormal logo pictures can be detected. Among them, the AR application can scan the logo picture and superimpose the model to actually recognize the feature point information on the logo picture. The more feature points, the easier it is to recognize. The feature point information on the same logo picture is fixed. For example, the AR SDK of Company A can quickly recognize the logo picture and complete the superposition of the model, while the AR SDK of Company B cannot recognize the logo picture after repeatedly adjusting the angle, indicating that the AR SDK of Company A is better at recognizing the feature points of the logo picture. Therefore, in order to evaluate the ability of the AR SDK to recognize logo pictures with different feature points, the logo pictures can be classified and stored in advance. In one embodiment, feature point information of the identification image is obtained, the identification image is classified according to the feature point information, and the classified identification image is stored. For example, the identification image is classified according to the size of the feature point data packet of the identification image, and the identification images with similar color values or color gamut features are classified into one category, and the identification images with obvious adjacent color block boundary features are classified into another category.
[0064] In one embodiment, when testing the AR application, the second test terminal will display the logo images under different test scenarios, wherein the display parameters of the logo images under different test scenarios are different. At this time, the AR application in the first test terminal is opened, and the camera is turned on to scan the logo images displayed by the second test terminal in different test scenarios, wherein when the camera is turned on to scan the logo images, continuous scanning can be achieved by recording a video. Then, the AR application identifies the scanned logo image to obtain feature point information. For example, a corresponding image scale space is generated based on the scanned logo image, and then the local extreme points in the image scale space are detected, and then the local extreme points are accurately located by eliminating low-contrast points and edge response points, and finally feature point information that can reflect the characteristics of the image is obtained.
[0065] In one embodiment, an AR application in a mobile phone is used to scan a Web 3D picture library application module on a browser side, and test scene instances in the Web 3D application are sequentially operated on the browser side, wherein the test scene includes:
[0066] Anti-light effect test, used to detect whether the AR SDK can be stably superimposed or the degree of jitter under strong light, weak light and dark light;
[0067] The FPS (Frames Per Second) value when the model is loaded on the terminal, which is used to detect the smoothness of ARSDK model loading;
[0068] Angle tilt test: automatically and slowly tilt the angle of the logo image to detect at which angles the AR SDK loses the model;
[0069] Comparative test of recognition graphs with different feature points, used to detect the recognition efficiency of AR SDK for different types of recognition graphs;
[0070] The near-far and far-field tests automatically zoom in and out on images to simulate scenes where the marker is near or far from the terminal. This is used to test the AR SDK model loading stability and target tracking stability.
[0071] Multi-target recognition capability test, used to detect the AR SDK's ability to track multiple targets to be identified;
[0072] The CPU and memory usage test is used to detect the CPU and memory usage of the terminal when loading the test model.
[0073] In different test scenarios, different identification pictures are displayed, and the mobile phone can identify the feature point information in different test scenarios by scanning the identification pictures.
[0074] Step S20, obtaining a model to be loaded according to the feature point information, wherein the recognition of the feature point information is realized by an AR software development kit to be detected;
[0075] After identifying the feature point information of the identification image, the model to be loaded is obtained according to the feature point information, wherein the model to be loaded refers to a pre-built three-dimensional model, that is, a 3D model (virtual model). When using the AR software development kit (AR SDK) to be detected to develop an AR application, the AR SDK to be detected is used to identify the feature point information of the identification image, complete the image registration, and at the same time, associate each identification image with a specific three-dimensional model. Therefore, after identifying the feature point information of the identification image, the feature point information is matched with the pre-stored feature point information to obtain a matching result. When the matching result is that the feature point information matches the pre-stored feature point information, the model to be loaded associated with the pre-stored feature point information is obtained. For example, when the AR application is opened or started, a pre-built three-dimensional model, such as a building, animation, animal, etc., is automatically extracted from the local application. At the same time, the pre-stored feature point information associated with the extracted three-dimensional model is obtained, and then the feature point information is matched with the pre-stored feature point information to obtain the associated model to be loaded.
[0076] Step S30, superimposing the model to be loaded with the feature point information to generate superimposed images under different test scenarios;
[0077] It should be noted that AR technology synthesizes scenes based on reality, adds some information to expand the data in the hands of users, applies virtual information to the real world, and combines the real environment with virtual objects to superimpose them in real time on the same space, scene, and screen.
[0078] After determining the model to be loaded, the AR application superimposes the model to be loaded with the feature point information to generate superimposed images under different test scenarios. For example, the AR application renders the three-dimensional model into a virtual image, and adjusts the shape and / or angle of the virtual image according to display requirements, and superimposes the adjusted virtual image with the real scene (feature point information, i.e., identification picture) to obtain a superimposed image.
[0079] Step S40: acquiring test parameters of each of the superimposed images, and evaluating the AR function of the AR application according to the test parameters.
[0080] When displaying the superimposed images, the test parameters of each superimposed image are obtained, and the AR function of the AR application is evaluated according to the test parameters. In one embodiment, the tester sets different weight values for the AR SDK's anti-light effect, the tracking ability of the marker, the multi-target recognition and positioning ability, the recognition ability of the marker with fewer feature points, and the FPS (Frames Per Second) and performance parameters of the AR SDK when loaded on the terminal. Then, according to the test parameters, the test scores of the AR function in different test scenarios are obtained, and the test scores are weighted and summed according to the weight values to obtain the target scores, and the AR function of the AR application is evaluated according to the target scores.
[0081] Optionally, different AR applications in multiple first test terminals can be tested. For example, each mobile phone is installed with an AR application developed based on a different AR SDK. The AR application of each mobile phone is used to simultaneously scan the logo image displayed by the second test terminal in different test scenarios, and then the superimposed images displayed on each mobile phone are compared. For example, in the same test scenario, the FPS value of AR SDK1 is 100 frames per second, the FPS value of AR SDK2 is 60 frames per second, and the FPS value of AR SDK3 is 30 frames per second. These values are statistically analyzed according to percentages to obtain the test score for the test scenario. Finally, the test score of each test scenario is weightedly calculated with the weight value of each test scenario to obtain the final test score of each AR SDK, and the AR function of each AR application is evaluated based on the test score.
[0082] This embodiment obtains feature point information by scanning the identification picture displayed by the second test terminal in different test scenarios. The display parameters of the identification picture in different test scenarios are different. The identification picture is a 3D graphic built based on the 3D drawing protocol WebGL and stored in the target library to be identified of the second test terminal; the model to be loaded is obtained according to the feature point information, and the identification of the feature point information is realized by the AR software development kit to be detected; the model to be loaded is superimposed with the feature point information to generate superimposed images in different test scenarios; the test parameters of each superimposed image are obtained, and the AR function of the AR application is evaluated according to the test parameters. This application performs multi-scenario testing on AR applications, obtains test parameters of superimposed images in different test scenarios, and evaluates the AR function of AR applications based on the test parameters, thereby improving the accuracy of AR testing.
[0083] Further, refer to Figure 3 , a second embodiment of the AR testing method of the present application is proposed.
[0084] The difference between the second embodiment of the AR testing method and the first embodiment is that the step of evaluating the AR function of the AR application according to the test parameters includes:
[0085] Step S41, obtaining weight values of different test scenarios;
[0086] Step S42, obtaining the test scores of the AR function in different test scenarios according to the test parameters;
[0087] Step S43: performing weighted summation on the test scores according to the weight values to obtain a target score, and evaluating the AR function of the AR application according to the target score.
[0088] It should be noted that in order to quantitatively evaluate the importance of different aspects of AR applications, it is necessary to treat the role of each test scenario in the overall evaluation differently, that is, to set a weight value for each test scenario to indicate the relative importance of the test scenario in the overall evaluation.
[0089] In one embodiment, the weight values of different test scenarios pre-set by the tester are obtained, and at the same time, the test scores of the AR function in different test scenarios are obtained according to the test parameters, wherein the corresponding test scoring rules and test parameters are also different due to the different types of test scenarios. When determining the type of the current test scenario, the test parameters of the test scenario are tested and scored for the AR function according to the test scoring rules corresponding to the test scenario, and the test scores of the AR function of the AR application in different test scenarios are obtained. For example, the test parameters of some test scenarios are data indicators, and then the scores are scored based on the data indicators, such as the less CPU resource occupancy, the better, that is, the less occupancy, the higher the test score; for another example, the larger the FPS (Frames PerSecond, the number of frames transmitted per second), the better, that is, the larger the FPS value, the higher the test score. And some test scenarios are tested and scored according to the actual loading of the model when scanning the picture. For example, under the same strong light interference, if the model is loaded quickly, 10 points are given, if the loaded model appears and disappears from time to time, 3 points are given, and if the model cannot be loaded, 0 points are given. For another example, when the marker has an inclination angle of 60 degrees, if the model can be loaded normally, 10 points will be given; if the model is lost or shakes abnormally, 0 points will be given.
[0090] After determining the test scores of the AR function in each test scenario, the test scores are weighted and summed according to the weight values to obtain the target score, and the AR function of the AR application is evaluated according to the target score. For example, the weight value of the lighting environment change test scenario is 10%, the test score is 60, the weight value of the left and right movement tracking test scenario is 20%, the test score is 50, the weight value of the angle tilt test scenario is 30%, the test score is 70, the weight value of the multi-target recognition test scenario is 40%, the test score is 80, then the target score is 10%*60+20%*50+30%*70+40%*80=69. Among them, the user pre-sets the AR function evaluation corresponding to different score levels in the AR application, for example, the target score ≥90 indicates that the AR function is very good; 70≤target score<90 indicates that the AR function is good; target score<70 indicates that the AR function is poor.
[0091] This embodiment obtains weight values under different test scenarios, and scores the test parameters for the AR function according to the test scoring rules corresponding to the test scenario type, and finally performs weighted summation of the test scores according to the weight values to obtain the target score, and evaluates the AR function of the AR application according to the target score. In this way, an overall evaluation can be performed on each test scenario to improve the accuracy of the evaluation results.
[0092] Further, refer to Figure 4 , a third embodiment of the AR testing method of the present application is proposed.
[0093] The difference between the third embodiment of the AR testing method and the second embodiment is that, before the step of scanning the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, the method includes:
[0094] Step S11, obtaining a pre-built three-dimensional model;
[0095] Step S12, identifying the logo image according to the AR software development kit to be detected;
[0096] Step S13: Bind the three-dimensional model to the identified logo image, and export an AR application package according to the binding information.
[0097] It should be noted that this application conducts a comprehensive evaluation of the AR SDK to be tested. Therefore, it is necessary to develop an AR application using the AR SDK to be tested, and then test the developed AR application.
[0098] In one embodiment, a pre-built three-dimensional model is obtained, wherein the rendering of the virtual three-dimensional model can be completed by 3D MAX, Blender, Maya or other 3D software tools, and an OBJ file that is relatively easy to load and process is exported. For example, a simple three-dimensional model (OBJ format, mtl material) is constructed using a lightweight 3D model software tool Blender. Then, the AR SDK and the logo image to be detected are obtained, and the feature points of the logo image are identified by the AR SDK to be detected, and image registration is completed. For example, based on a deep learning method, the relative position of the three-dimensional object relative to the device is determined by three-dimensional object detection, and the coordinate relationship between the virtual space and the real space is established. Among them, image registration refers to the process of establishing the coordinate relationship between the virtual space and the real space based on the recognition information after the scene or object is identified. As a key technology in augmented reality technology, image registration technology directly affects the superposition effect and visualization effect of virtual information in the real scene. Finally, the constructed 3D model and logo image are combined in Unity 3D, and the functional interaction effect of the model is completed at the same time. The constructed virtual and real scene is exported as an APK package, and exported to a folder, installed on a mobile device, and AR is displayed on a mobile device.
[0099] This embodiment builds a three-dimensional model, completes image registration based on the AR SDK to be tested and the identification picture, then completes the construction of the virtual and real scenes based on the three-dimensional model and the identification picture, and exports the APK package. In this way, the test terminal can evaluate the function of the AR SDK by testing the AR application installed based on the APK package.
[0100] In order to better illustrate the AR testing method of this application, refer to Figure 5 , Figure 5 This is an operational flow chart of the AR testing method of this application.
[0101] This embodiment proposes an AR testing system, which includes a logo recognition picture library module, a virtual and real scene construction module, a mobile terminal output module, and an augmented reality effect indicator statistics module.
[0102] The logo recognition image library module is used to store the target image library to be recognized. Its feature is that it uses WebGL technology to build Web 3D applications. Since it does not require the installation of any browser plug-ins, the loading, display and scene interaction of 3D scenes can be completed through WebGL technology and related open source frameworks. Therefore, the logo recognition image library module can not only store a large number of logo images to be recognized, but also automatically simulate interference operations on the images to be recognized during testing based on the interactive capabilities of WebGL, such as large inclination angles, long and short distances, abnormal lighting, etc. of the images to be recognized. These operations can be achieved by setting information parameters to avoid unfair testing problems caused by environmental errors.
[0103] The virtual-reality scene building module is used to complete the superposition of virtual 3D models or animations with the pictures to be identified. This module creates multiple object database files by superimposing virtual scenes (such as 3D models, animations or other virtual information) with real scenes using the AR SDK to be detected. Specifically, the virtual model is rendered through 3DMAX, Blender, Maya or other 3D software tools, and the OBJ file that is easier to load and process is exported. Then the virtual model and the picture to be identified are imported into Unity 3D and the image registration of the picture to be identified is completed using the AR SDK to be detected.
[0104] The mobile terminal output module is used to complete the loading, display and functional interaction of virtual and real scenes on mobile terminal devices. Based on the completion of virtual and real construction synthesis and image registration, when the mobile terminal device is used to scan the identification picture, the module loads and displays the virtual information on the terminal device and can perform human-computer interaction operations on it.
[0105] The augmented reality effect indicator statistics module is used to complete the statistics of the augmented reality effect of the AR SDK to be tested under various scene tests, including the jitter and offset effects of the AR engine when loading the model, the motion tracking ability of the identified object, the ability to resist abnormal lighting interference, memory and CPU usage, FPS parameters, multi-target recognition ability, etc. Among them, the tester can set different weight values for each test scene according to actual needs, and finally perform weighted calculation based on the weight value to obtain the total score of the AR engine, and evaluate the AR SDK to be tested based on the total score.
[0106] This embodiment fills the gap in the systematic functional testing of AR SDK to a certain extent. It uses WebGL technology to build a Web 3D application to simulate abnormal operation scenarios of markers, avoiding the imprecision of manual operation testing. At the same time, based on the highly operable marker recognition picture library module, through multi-scenario testing and obtaining various performance results of the AR SDK, the AR SDK is evaluated from a multi-faceted functional level without paying too much attention to the underlying AR technology. At the same time, this embodiment does not require the use of complex graphics and image processing equipment, and has the characteristics of ease of use, reliability and high reusability.
[0107] In addition, the present application also provides an AR testing device, which includes a memory, a processor, and an AR testing program stored in the memory and running on the processor, and when the processor executes the AR testing program, the steps of the AR testing method described above are implemented.
[0108] refer to Figure 6 The AR testing device 100 includes a scanning module 10, an acquisition module 20, a superposition module 30 and an evaluation module 40, wherein:
[0109] The scanning module 10 is used to scan the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, the display parameters of the identification picture in different test scenarios are different, the identification picture is a 3D graphic constructed based on the 3D drawing protocol WebGL, and is stored in the target library to be identified of the second test terminal;
[0110] The acquisition module 20 is used to acquire the model to be loaded according to the feature point information, and the recognition of the feature point information is realized by the AR software development kit to be detected;
[0111] The superposition module 30 is used to superimpose the model to be loaded with the feature point information to generate superimposed images under different test scenarios;
[0112] The evaluation module 40 is used to obtain test parameters of each of the superimposed images, and evaluate the AR function of the AR application according to the test parameters.
[0113] In addition, the present application also provides a computer-readable storage medium, on which an AR testing method program is stored. When the AR testing method program is executed by a processor, the steps of the above AR testing method are implemented.
[0114] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0115] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0116] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0117] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0118] It should be noted that in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The present application may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.
[0119] Although the optional embodiments of the present application have been described, those skilled in the art, once informed of the basic creative concept, may make additional changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including optional embodiments and all changes and modifications falling within the scope of the present application.
[0120] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims
1. An AR testing method, characterized in that: The AR testing method is applied to a first test terminal, and the method includes: Scanning the identification pictures displayed by the second test terminal in different test scenarios to obtain feature point information, wherein the display parameters of the identification pictures in the different test scenarios are different, and the identification pictures are 3D graphics constructed based on the 3D drawing protocol WebGL and stored in the target library to be identified of the second test terminal; Matching the feature point information with pre-stored feature point information to obtain a matching result; When the matching result is that the feature point information matches the pre-stored feature point information, obtaining the model to be loaded associated with the pre-stored feature point information, wherein the feature point information is associated with the model to be loaded in advance, and the recognition of the feature point information is realized by the AR software development kit to be detected; Rendering the model to be loaded into a virtual image; Superimposing the virtual image and the feature point information to generate superimposed images under different test scenarios; Acquire test parameters of each of the superimposed images, and evaluate the AR function of the AR application according to the test parameters.
2. The AR testing method according to claim 1, characterized in that: The step of evaluating the AR function of the AR application according to the test parameters comprises: Get the weight values of different test scenarios; Acquire test scores of the AR function in different test scenarios according to the test parameters; The test scores are weighted and summed according to the weight values to obtain a target score, and the AR function of the AR application is evaluated according to the target score.
3. The AR testing method according to claim 2, characterized in that: The step of obtaining the test scores of the AR function in different test scenarios according to the test parameters includes: Obtaining test scoring rules and test parameters corresponding to the type of each test scenario; Under each type of the test scenario, the test parameters of the test scenario are scored for AR function testing according to the test scoring rules corresponding to the test scenario to obtain test scores of the AR function of the AR application under different test scenarios.
4. The AR testing method according to claim 2, characterized in that: The step of evaluating the AR function of the AR application according to the target score comprises: Obtaining the score level corresponding to the target score; The AR function of the AR application is evaluated according to the score level.
5. The AR testing method according to claim 1, characterized in that: Before the step of scanning the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, the method includes: Get pre-built 3D models; Identify the logo image according to the AR software development kit to be detected; The three-dimensional model is bound to the identified logo image, and an AR application package is exported according to the binding information.
6. An AR testing device, characterized in that: The device includes a memory, a processor, and an AR test program stored in the memory and running on the processor, and the processor implements the steps of the method according to any one of claims 1 to 5 when executing the AR test program.
7. An AR testing device, characterized in that: The device comprises a scanning module, an acquisition module, a superposition module and an evaluation module, wherein: The scanning module is used to scan the identification picture displayed by the second test terminal in different test scenarios to obtain feature point information, the display parameters of the identification picture in different test scenarios are different, the identification picture is a 3D graphic constructed based on the 3D drawing protocol WebGL, and is stored in the target library to be identified of the second test terminal; The acquisition module is used to match the feature point information with the pre-stored feature point information to obtain a matching result; when the matching result is that the feature point information matches the pre-stored feature point information, the model to be loaded associated with the pre-stored feature point information is acquired, wherein the feature point information is associated with the model to be loaded in advance, and the recognition of the feature point information is realized by the AR software development kit to be detected; The superposition module is used to render the model to be loaded into a virtual image; superimpose the virtual image with the feature point information to generate superimposed images under different test scenarios; The evaluation module is used to obtain test parameters of each of the superimposed images and evaluate the AR function of the AR application according to the test parameters.
8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an AR test program, and when the AR test program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.
