Augmented Reality-based Feature Point Cloud Stereo Object Framing Method and Device

Abstract

The embodiments of the present invention disclose a method and device for stereoscopic object bounding of feature point clouds based on augmented reality. In the embodiments, video data of a target object is collected, and feature point cloud data is generated based on the video data, thereby establishing a corresponding coordinate system, searching for target feature point clouds and obtaining the coordinate values of the target feature point clouds. For the feature point clouds whose coordinate values based on the target feature point clouds are too far from the target object, since the feature point cloud data has been screened, it is possible to focus on the feature point cloud cluster of the target object to be measured, and at the same time, after removing the feature point clouds that are too far away, it can also help the user determine whether the object actually wanted to be measured by the user has been selected, thereby helping to accurately bound the target object, and further enabling accurate volume calculation of the target object.

Description

Technical Field

[0001] The present invention relates to the field of virtual reality technology, and particularly to a method and device for frame selection of a three-dimensional object based on feature point cloud in augmented reality. Background Art

[0002] Currently, with the convenience brought by the development of the Internet and the popularity of mobile devices, users can easily obtain video stream information of the real world. SLAM (simultaneous localization and mapping), also known as CML (Concurrent Mapping and Localization), real-time localization and mapping, or concurrent mapping and localization, is mainly used to solve the problems of positioning, navigation, and mapping of mobile robots running in unknown environments. Using the SLAM algorithm to process video stream information to model the real environment is a basic method in AR augmented reality technology. However, the accuracy of the SLAM algorithm can only roughly model the real world, and it is difficult to achieve some operations that require fine processing, such as accurately positioning an object and calculating the volume of an object. Summary of the Invention

[0003] In view of the above deficiencies, embodiments of the present invention disclose a method, device, electronic device, and storage medium for frame selection of a three-dimensional object based on feature point cloud in augmented reality, which can solve the deficiencies of the prior art that cannot accurately locate an object and calculate its volume, and achieve high-precision frame selection of an object.

[0004] The first aspect of the embodiments of the present invention discloses a method for frame selection of a three-dimensional object based on feature point cloud in augmented reality, including:

[0005] Collect video data of a target object and generate feature point cloud data of the video data;

[0006] Establish a coordinate system of the target object based on the feature point cloud data, and obtain the coordinate values of the feature point cloud of the target object in the coordinate system;

[0007] Define the feature point cloud corresponding to each boundary of the target object as the target feature point cloud, and based on the coordinate values of the target feature point cloud in the coordinate system, exclude the feature point cloud outside the target object and at a distance greater than a first threshold from the target feature point cloud from all the feature point cloud data.

[0008] As an optional implementation manner, in the first aspect of the embodiments of the present invention, the collecting video data of the target object includes:

[0009] Collect video data of the target object from different angles through a handheld camera.

[0010] As an alternative implementation, in the first aspect of the embodiments of the present invention, the step of collecting video data of the target object from different angles by the handheld camera includes:

[0011] Configuring video acquisition calibration data and displaying the calibration data when a video acquisition instruction input by the user is received;

[0012] Detecting whether the video frame captured by the current handheld camera is consistent with the calibration data;

[0013] When the video frame is consistent with the calibration data, collecting video data of the target object by the handheld camera.

[0014] As an alternative implementation, in the first aspect of the embodiments of the present invention, the step of establishing a coordinate system of the target object based on the feature point cloud data and obtaining the coordinate values of the feature point cloud of the target object in the coordinate system includes:

[0015] Selecting the center point of the video frame as the origin of the coordinate system and establishing a coordinate system of the target object based on the feature point cloud data;

[0016] Obtaining the coordinate values of the feature point cloud of the target object in the coordinate system.

[0017] As an alternative implementation, in the first aspect of the embodiments of the present invention, the step of defining the feature point cloud corresponding to each boundary of the target object as the target feature point cloud includes:

[0018] Taking the center point of the video frame as the origin, generating virtual rays with the origin as the end points; the virtual rays intersect with each boundary of the target object in the video frame to obtain a number of intersection points of the virtual rays and the video frame, forming an intersection point set.

[0019] As an alternative implementation, in the first aspect of the embodiments of the present invention, it further includes:

[0020] Traversing and removing the feature point clouds outside the target object and having a distance greater than the first threshold from the target feature point cloud, and defining the feature point cloud as the target data set;

[0021] Selecting the feature point clouds within the second threshold from the target data set to construct a new point cloud data set.

[0022] As an alternative implementation, in the first aspect of the embodiments of the present invention, the step of selecting the feature point clouds within the second threshold from the target data set to construct a new point cloud data set includes:

[0023] Select any target feature point, and screen out the point cloud of feature points within the second threshold from the target dataset as the new point cloud of feature points;

[0024] Construct a new point cloud dataset, determine whether the number of the new point cloud of feature points is greater than a preset value, and when the omission of the new point cloud of feature points is greater than the preset value, add the new point cloud of feature points to the new point cloud dataset;

[0025] Repeat the above steps until all target feature points are selected.

[0026] The second aspect of the embodiments of the present invention discloses a stereoscopic object frame selection device for feature point cloud based on augmented reality, including:

[0027] Data acquisition module: used to acquire the video data of the target object and generate the feature point cloud data of the video data;

[0028] Coordinate system establishment module: used to establish the coordinate system of the target object based on the feature point cloud data and obtain the coordinate values of the feature point cloud of the target object in the coordinate system;

[0029] Point cloud frame selection module: used to define the feature point cloud corresponding to each boundary of the target object as the target feature point cloud, and based on the coordinate values of the target feature point cloud in the coordinate system, exclude the feature point cloud outside the target object and with a distance greater than the first threshold from all the feature point cloud data.

[0030] As an optional implementation manner, in the second aspect of the embodiments of the present invention, in the data acquisition module, the acquisition of the video data of the target object includes:

[0031] Acquire the video data of the target object from different angles through a handheld camera.

[0032] As an optional implementation manner, in the second aspect of the embodiments of the present invention, the acquisition of the video data of the target object from different angles through a handheld camera includes:

[0033] Configure video acquisition calibration data, and when receiving a user input video acquisition instruction, display the calibration data;

[0034] Detect whether the video picture collected by the current handheld camera is consistent with the calibration data;

[0035] When the video picture is consistent with the calibration data, acquire the video data of the target object through the handheld camera.

[0036] As an alternative implementation manner, in the second aspect of the embodiments of the present invention, in the coordinate system establishment module, establishing a coordinate system of the target object based on the feature point cloud data and obtaining the coordinate values of the feature point cloud of the target object in the coordinate system includes:

[0037] Selecting the center point of the video frame as the origin of the coordinate system and establishing a coordinate system of the target object based on the feature point cloud data;

[0038] Obtaining the coordinate values of the feature point cloud of the target object in the coordinate system.

[0039] As an alternative implementation manner, in the second aspect of the embodiments of the present invention, in the point cloud selection module, defining the feature point cloud corresponding to each boundary of the target object as the target feature point cloud includes:

[0040] Taking the center point of the video frame as the origin to generate virtual rays with the origin as the end points; each virtual ray intersects with each boundary of the target object in the video frame to obtain a number of intersection points of the virtual rays and the video frame, forming an intersection point set.

[0041] As an alternative implementation manner, in the second aspect of the embodiments of the present invention, it further includes a point cloud elimination module: for traversing and eliminating the feature point cloud outside the target object and having a distance greater than a first threshold from the target feature point cloud, and defining this feature point cloud as the target data set; screening out the feature point cloud within a second threshold from the target feature point cloud in the target data set to construct a new point cloud data set.

[0042] As an alternative implementation manner, in the second aspect of the embodiments of the present invention, screening out the feature point cloud within a second threshold from the target feature point cloud in the target data set to construct a new point cloud data set includes:

[0043] Selecting any one target feature point and screening out the feature point cloud within a second threshold from this target feature point in the target data set as the new feature point cloud;

[0044] Constructing a new point cloud data set, judging whether the number of the new feature point cloud is greater than a preset value, and when the omission of the new feature point cloud is greater than the preset value, adding this new feature point cloud to the new point cloud data set;

[0045] Repeating the above steps until all target feature points are selected.

[0046] The third aspect of the embodiments of the present invention discloses an electronic device, including: a memory storing executable program code; a processor coupled to the memory; the processor calling the executable program code stored in the memory for executing the method for stereoscopic object selection of feature point cloud based on augmented reality disclosed in the first aspect of the embodiments of the present invention.

[0047] The fourth aspect of the embodiments of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method for stereoscopic object bounding of feature point clouds based on augmented reality disclosed in the first aspect of the embodiments of the present invention.

[0048] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0049] In the embodiments of the present invention, video data of a target object is collected, feature point cloud data is generated based on the video data, a corresponding coordinate system is established, target feature point clouds are searched for and the coordinate values of the target feature point clouds are obtained. For feature point clouds whose coordinate values based on the target feature point clouds are too far from the target object, since the feature point cloud data has been screened, it is possible to focus on the cluster of feature point clouds of the target object to be measured, and at the same time, after removing the feature point clouds that are too far away, it can also help the user determine whether the object actually wanted to be measured by the user has been selected, thereby helping to accurately bound the target object, and further enabling accurate volume calculation of the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0051] Figure 1 is a schematic flowchart of a method for stereoscopic object bounding of feature point clouds based on augmented reality provided by the embodiments of the present invention;

[0052] Figure 2 is a schematic structural diagram of a device for stereoscopic object bounding of feature point clouds based on augmented reality provided by the embodiments of the present invention;

[0053] Figure 3 is a schematic structural diagram of an electronic device provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0055] It should be noted that the terms "first", "second", "third", "fourth", etc. in the description and claims of the present invention are used to distinguish different objects, rather than to describe a specific order. The terms "comprising" and "having" in the embodiments of the present invention and any variations thereof are intended to cover non-exclusive inclusion. Exemplarily, a process, method, system, product or device comprising a series of steps or units need not be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0056] The embodiments of the present invention disclose a method, device, electronic device and storage medium for stereoscopic object bounding of feature point clouds based on augmented reality. In the embodiments, video data of a target object is collected, and feature point cloud data is generated based on the video data, thereby establishing a corresponding coordinate system, finding target feature point clouds and obtaining the coordinate values of the target feature point clouds. Feature point clouds that are too far from the target object based on the coordinate values of the target feature point clouds can be focused on the cluster of feature point clouds of the target object to be measured after screening the feature point cloud data, and at the same time, the removal of feature point clouds that are too far can also help the user determine whether the object actually wanted to be measured by the user is selected, thereby helping to accurately bound the target object, and further enabling accurate volume calculation of the target object.

[0057] Embodiment 1

[0058] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of the method for stereoscopic object bounding of feature point clouds based on augmented reality disclosed in the embodiments of the present invention. Among them, the execution subject of the method described in the embodiments of the present invention is an execution subject composed of software or / and hardware. The execution subject can receive relevant information through wired or / and wireless means and can send certain instructions. Of course, it can also have certain processing functions and storage functions. The execution subject can control multiple devices, such as remote physical servers or cloud servers and related software, or can also be a local host or server and related software that performs relevant operations on devices placed somewhere. In some scenarios, multiple storage devices can also be controlled, and the storage devices can be placed in the same place or different places as the devices. As Figure 1 shown, the method for stereoscopic object bounding of feature point clouds based on augmented reality includes the following steps:

[0059] S101: Collect video data of a target object and generate feature point cloud data of the video data.

[0060] In the embodiment, the target object is also the object for which video data needs to be collected currently. For example, it can be a suitcase. When the user needs to perform contour bounding on any object, this object serves as the target object, and the user first collects the video data of this target object. The pixels of the video data are not specifically limited in the embodiment, as long as the actual requirements are met.

[0061] Specifically, a handheld camera is used to collect the video data of the target object. This handheld camera can be a dedicated camera device or the intelligent camera built into an intelligent device, such as a mobile phone, a smart tablet, a laptop computer, etc. The intelligent devices in the above examples usually have a photography function in addition to their intelligent functions, and a camera is set on their bodies. Through this camera, video calls, image acquisition, and video acquisition can be performed. Since these intelligent devices can all be moved while held by the user, they are also called mobile cameras.

[0062] In the embodiment, in order to collect rich and complete video data of the target object, the user can collect the video data of the target object from different angles through the handheld camera. In practical applications, an application program can be set. When the user needs to perform bounding on the target object, the application program is triggered, and the camera is called in the application program to collect video data. At this time, the application program prompts the user to collect video data according to the corresponding photographing angles. Specifically, video acquisition calibration data is configured, and when a video acquisition instruction input by the user is received, the calibration data is displayed; it is detected whether the video picture collected by the current handheld camera is consistent with the calibration data; when the video picture is consistent with the calibration data, the video data of the target object is collected through the handheld camera.

[0063] In the above, configuring the video acquisition calibration data means that a photographing angle comparison benchmark is configured in advance in the application program. This comparison benchmark can be displayed in the form of an image on the screen interface. When the user determines to start video acquisition, the calibration data is displayed, that is, the user can intuitively compare whether the currently collected target item falls in the appropriate position through the display of the image.

[0064] S102: Establish a coordinate system of the target object based on the feature point cloud data, and obtain the coordinate values of the feature point cloud of the target object in the coordinate system.

[0065] After the embodiment collects the video data and obtains the feature point cloud data based on the SLAM algorithm, the SLAM algorithm is then used to establish the coordinate system of the target object in the video data. Since the coordinate values of each pixel point in the coordinate system can be obtained, it is convenient to screen the feature points in subsequent operations.

[0066] Specifically, first select the origin of the coordinate system, and then establish a coordinate system based on the origin of the coordinate system. As an example, the embodiment can select the center point of the video screen as the origin of the coordinate system, and establish the coordinate system of the target object based on the feature point cloud data. Based on this coordinate system, the coordinate values of the feature points of the target object in the coordinate system are obtained. In the embodiment, video data can be obtained at different angles, that is, the video stream is transformed, and the SLAM algorithm is continuously used to refine the virtual coordinate system until the virtual coordinate system is stable. The user can be reminded to scan the suitcase from different angles, and real-world information can be obtained from different angles to assist the SLAM algorithm in obtaining a stable virtual coordinate system, and the virtual feature point cloud corresponding to the real world and the virtual plane information where the target item is placed can be obtained through the SLAM algorithm.

[0067] S103: Define the feature point cloud corresponding to each boundary of the target object as the target feature point cloud, and based on the coordinate values of the target feature point cloud in the coordinate system, exclude the feature point cloud outside the target object and with a distance greater than the first threshold from the target feature point cloud from all the feature point cloud data.

[0068] When guiding the user to start the calculation, the camera is aligned with the suitcase. During the calculation, the position of the center point of the user's mobile phone screen in the virtual coordinate system is used as the starting point. In the embodiment, with the center point of the video screen as the origin, a virtual ray with the origin as the endpoint is generated; the virtual ray intersects each boundary of the target object in the video screen to obtain several intersection points of the virtual ray and the video screen, forming an intersection point set.

[0069] Preferably, the embodiment can further include traversing and excluding the feature point cloud outside the target object and with a distance greater than the first threshold from the target feature point cloud, and defining this feature point cloud as the target data set; screening out the feature point cloud within the second threshold from the target data set to construct a new point cloud data set. On the basis of forming the framework of the target object, the embodiment further excludes the feature point cloud that is too far away from the target object, so the remaining feature points are all closer to the target object, that is, the formed framework of the target object is more accurate. The first threshold is, for example, 0.05 units, and is specifically set by the operating device, compiler, and other actual situations.

[0070] Among the above, filtering out the feature point clouds within the second threshold from the target data set to construct a new point cloud data set includes: selecting any one target feature point, and filtering out the feature point clouds within the second threshold from the target data set as the new feature point clouds; constructing a new point cloud data set, determining whether the number of the new feature point clouds is greater than a preset value, and when the omission of the new feature point clouds is greater than the preset value, adding the new feature point clouds to the new point cloud data set; repeating the above steps until all the target feature points are selected. The second threshold is, for example, 0.03 units, and the preset value is, for example, 3. In the embodiment, for example, there are five target feature points. The above steps are sequentially executed for each target feature point. One of the target feature points is selected as the central target feature point, and other feature points including the central target feature point and with a distance less than or equal to 0.03 are found. When the number of other feature points plus the central target feature point is greater than 3, the new feature point clouds of the central target feature point are formed and added to the new point cloud data set. Based on this process, the non-aggregated feature points can be removed, and more accurate framing can be performed, and then the volume data of the target object can be obtained accurately.

[0071] Embodiment 2

[0072] Please refer to Figure 2 , Figure 2 which is a schematic structural diagram of an augmented reality-based feature point cloud three-dimensional object framing device disclosed in an embodiment of the present invention. As Figure 2 shown, the augmented reality-based feature point cloud three-dimensional object framing device may include: a data acquisition module 201, a coordinate system establishment module 202, and a point cloud framing module 203. Among them, the data acquisition module 201 is used to acquire video data of a target object and generate feature point cloud data of the video data; the coordinate system establishment module 202 is used to establish a coordinate system of the target object based on the feature point cloud data and obtain the coordinate values of the feature point cloud of the target object in the coordinate system; the point cloud framing module 203 is used to define the feature point cloud corresponding to each boundary of the target object as the target feature point cloud, and based on the coordinate values of the target feature point cloud in the coordinate system, filter out the feature point clouds outside the target object and with a distance greater than the first threshold from all the feature point cloud data.

[0073] In the above data acquisition module 201, the acquisition of the video data of the target object includes: acquiring the video data of the target object from different angles through a hand-held camera. Further, the acquisition of the video data of the target object from different angles through a hand-held camera includes: configuring video acquisition calibration data, and when a video acquisition instruction input by the user is received, displaying the calibration data; detecting whether the video picture acquired by the current hand-held camera is consistent with the calibration data; when the video picture is consistent with the calibration data, acquiring the video data of the target object through the hand-held camera.

[0074] In the above coordinate system establishment module 202, the establishment of the coordinate system of the target object based on the feature point cloud data and the acquisition of the coordinate values of the feature point cloud of the target object in the coordinate system include: selecting the center point of the video picture as the origin of the coordinate system, and establishing the coordinate system of the target object based on the feature point cloud data; acquiring the coordinate values of the feature point cloud of the target object in the coordinate system.

[0075] Furthermore, in the point cloud bounding module 203, the definition of the feature point cloud corresponding to each boundary of the target object as the target feature point cloud includes: generating a virtual ray with the center point of the video picture as the origin; the virtual ray intersects with each boundary of the target object in the video picture to obtain a plurality of intersection points of the virtual ray and the video picture, forming an intersection point set.

[0076] In other examples, a point cloud rejection module may further be included: for traversing and rejecting the feature point clouds outside the target object and having a distance greater than a first threshold from the target feature point cloud, and defining the feature point cloud as a target data set; screening out the feature point clouds within a second threshold from the target feature point cloud in the target data set to construct a new point cloud data set. Among them, screening out the feature point clouds within a second threshold from the target feature point cloud in the target data set to construct a new point cloud data set includes: selecting any one target feature point, screening out the feature point clouds within a second threshold from the target feature point in the target data set as new feature point clouds; constructing a new point cloud data set, judging whether the number of the new feature point clouds is greater than a preset value, and when the omission of the new feature point clouds is greater than the preset value, adding the new feature point clouds to the new point cloud data set; repeating the above steps until all target feature points are selected.

[0077] Embodiment III

[0078] Please refer to Figure 3 , Figure 3It is a schematic structural diagram of an electronic device disclosed in an embodiment of the present invention. The electronic device can be a computer, a server, etc. Of course, in certain cases, it can also be an intelligent device such as a mobile phone, a tablet computer, and a monitoring terminal, as well as an image acquisition device with processing functions. As shown in the figure.., the electronic device may include:

[0079] A memory 301 storing executable program code;

[0080] A processor 302 coupled to the memory 301;

[0081] Wherein, the processor 302 calls the executable program code stored in the memory 301 and executes some or all of the steps in the method for stereoscopic object bounding of feature point cloud based on augmented reality in the first embodiment.

[0082] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute some or all of the steps in the method for stereoscopic object bounding of feature point cloud based on augmented reality in the first embodiment.

[0083] An embodiment of the present invention further discloses a computer program product, wherein when the computer program product runs on a computer, it causes the computer to execute some or all of the steps in the method for stereoscopic object bounding of feature point cloud based on augmented reality in the first embodiment.

[0084] An embodiment of the present invention further discloses an application publishing platform, wherein the application publishing platform is used to publish a computer program product, and when the computer program product runs on a computer, it causes the computer to execute some or all of the steps in the method for stereoscopic object bounding of feature point cloud based on augmented reality in the first embodiment.

[0085] In various embodiments of the present invention, it should be understood that the magnitude of the sequence numbers of the various processes does not necessarily mean the order of execution. The order of execution of the various processes should be determined by their functions and internal logics, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0086] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0087] In addition, in each embodiment of the present invention, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0088] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc., specifically, the processor in the computer device) to execute some or all of the steps of the methods described in the various embodiments of the present invention.

[0089] In the embodiments provided by the present invention, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information.

[0090] Those of ordinary skill in the art can understand that some or all of the steps in the various methods of the embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electrically-erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc memories, magnetic disk memories, tape memories, or any other computer-readable medium capable of carrying or storing data.

[0091] The above has introduced in detail the method, apparatus, electronic device and storage medium for frame selection of feature point cloud stereo objects based on augmented reality. In this article, specific examples are used to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for stereoscopic object bounding of feature point clouds based on augmented reality, characterized in that, Including: Collecting video data of a target object and generating feature point cloud data of the video data; Establishing a coordinate system of the target object based on the feature point cloud data and obtaining the coordinate values of the feature point cloud of the target object in the coordinate system; Defining the feature point cloud corresponding to each boundary of the target object as the target feature point cloud, and based on the coordinate values of the target feature point cloud in the coordinate system, removing the feature point cloud outside the target object and with a distance greater than a first threshold from the target feature point cloud from all the feature point cloud data.

2. The method for stereoscopic object bounding of feature point cloud based on augmented reality according to claim 1, wherein The collecting of the video data of the target object includes: Collecting the video data of the target object from different angles through a handheld camera.

3. The method for stereoscopic object bounding of feature point cloud based on augmented reality according to claim 2, characterized in that, The collecting of the video data of the target object from different angles through a handheld camera includes: Configuring video acquisition calibration data and displaying the calibration data when receiving a video acquisition instruction input by the user; Detecting whether the video frame collected by the current handheld camera is consistent with the calibration data; When the video frame is consistent with the calibration data, collecting the video data of the target object through the handheld camera.

4. The method for stereoscopic object bounding of feature point cloud based on augmented reality according to claim 3, wherein The establishing of the coordinate system of the target object based on the feature point cloud data and obtaining the coordinate values of the feature point cloud of the target object in the coordinate system includes: Selecting the center point of the video frame as the origin of the coordinate system and establishing the coordinate system of the target object based on the feature point cloud data; Obtaining the coordinate values of the feature point cloud of the target object in the coordinate system.

5. The method for stereoscopic object bounding of feature point clouds based on augmented reality according to claim 4, characterized in that, The defining of the feature point cloud corresponding to each boundary of the target object as the target feature point cloud includes: Taking the center point of the video frame as the origin and generating virtual rays with the origin as the end points; the virtual rays intersect with each boundary of the target object in the video frame to obtain a number of intersection points of the virtual rays and the video frame, forming an intersection point set.

6. The method for stereoscopic object bounding of feature point cloud based on augmented reality according to claim 5, wherein Also including: Traversing and removing the feature point cloud outside the target object and with a distance greater than a first threshold from the target feature point cloud, and defining the feature point cloud as the target data set; Selecting the feature point cloud within a second threshold from the target feature point cloud from the target data set to construct a new point cloud data set.

7. The method for stereoscopic object bounding of feature point clouds based on augmented reality according to claim 6, wherein The selecting of the feature point cloud within a second threshold from the target feature point cloud from the target data set to construct a new point cloud data set includes: Selecting any one target feature point and selecting the feature point cloud within a second threshold from the target data set from this target feature point as the new feature point cloud; Constructing a new point cloud data set, judging whether the number of the new feature point cloud is greater than a preset value, and when the omission of the new feature point cloud is greater than the preset value, adding the new feature point cloud to the new point cloud data set; Repeating the above steps until all the target feature points are selected.

8. An augmented reality-based feature point cloud stereo object framing device, characterized in that, Including: Data acquisition module: used for collecting the video data of the target object and generating the feature point cloud data of the video data; Coordinate system establishment module: used for establishing the coordinate system of the target object based on the feature point cloud data and obtaining the coordinate values of the feature point cloud of the target object in the coordinate system; Point cloud bounding box module: used to define the feature point cloud corresponding to each boundary of the target object as the target feature point cloud, and based on the coordinate values of the target feature point cloud in the coordinate system, remove the feature point cloud outside the target object and with a distance greater than the first threshold from the target feature point cloud from all the feature point cloud data.

9. An electronic device, characterized in that, Comprising: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and is used to execute the augmented reality-based feature point cloud three-dimensional object bounding method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program causes the computer to execute the augmented reality-based feature point cloud three-dimensional object bounding method according to any one of claims 1 to 7.

Images