Spatial pose anchoring method and device and XR equipment

Abstract

The embodiment of the invention discloses a space pose anchoring method and device and XR equipment, relates to the technical field of computer vision, and is used for solving the problems that in an existing space pose anchoring mode, it is difficult to accurately achieve global expression of the pose of a specific object, and anchoring references of multiple pieces of equipment are not uniform. The method comprises the following steps: acquiring current frame image data acquired by a sensor, and matching the current frame image data with target image data to obtain homonymy point pairs; according to the coordinate system of the target object and the homonymy point pair, solving the spatial pose of the current frame of image data to realize the anchoring of the spatial pose; wherein the target coordinate system is obtained by constructing the target image data based on the intelligent equipment in the current environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer vision, and in particular to a spatial pose anchoring method and device and an XR device. BACKGROUND

[0002] With the development of science and technology, in many technical fields such as augmented reality, virtual reality and mixed reality, it is necessary to interact between virtual and real in the actual scene, and to accurately superimpose virtual objects on specific positions in the real world, which requires accurate knowledge of the pose of the device relative to the real scene. Therefore, spatial pose anchoring of intelligent devices is an important link in many fields.

[0003] Existing intelligent devices mainly estimate the state by combining multiple sensors to obtain the position, speed, attitude and other information of the device carrier. For example, the carrier state information obtained by SLAM technology, but the reference system of this information is a local coordinate system. On the one hand, this coordinate system is not suitable for global expression of the attitude of a specific object, and on the other hand, in the case of loss of carrier positioning information, the current reference system needs to be repositioned to a specific global coordinate system to express all states uniformly. The visual odometry based on SLAM, GPS positioning and other methods often rely on global environmental features. This dependence can be disturbed by the natural environment, resulting in low positioning accuracy and ineffective anchoring. In addition, existing multiple intelligent devices may construct independent coordinate systems based on their own SLAM, resulting in the problem of inconsistent coordinate systems in the pose calibration between existing multiple intelligent devices, which makes the overall consistency between multiple devices low. SUMMARY

[0004] In view of the problem in the prior art that it is difficult to accurately achieve global expression of the attitude of a specific object and the anchoring reference of multiple devices is not unified, one or more embodiments of the present application provide a spatial pose anchoring method, device and XR device.

[0005] One or more embodiments of the present application adopt the following technical solutions: In a first aspect, one or more embodiments of the present application provide a spatial pose anchoring method, which comprises: obtaining current frame image data collected by a sensor, and matching the current frame image data with target image data to obtain corresponding point pairs; solving the spatial pose of the current frame image data according to the target object coordinate system and the corresponding point pairs, to achieve spatial pose anchoring; wherein the target coordinate system is obtained by constructing the target image data by the intelligent device in the current environment.

[0006] Optionally, in one or more embodiments of the present application, according to the target object coordinate system and the homonym point pair, the spatial pose of the current frame image data is solved to realize spatial pose anchoring, specifically including: According to the construction rule of the target object coordinate system, the current frame image data in the homonym point is mapped to obtain a matching relationship of the homonym point pair; Based on the matching relationship of the homonym point pair, the spatial pose transformation relationship of the current frame image data relative to the target object coordinate system is obtained to perform spatial pose anchoring based on the spatial pose transformation relationship.

[0007] Optionally, in one or more embodiments of the present application, based on the matching relationship of the homonym point pair, the spatial pose transformation relationship of the current frame image data relative to the target object coordinate system is obtained to perform spatial pose anchoring based on the spatial pose transformation relationship, specifically including: According to the matching relationship and the sensor internal parameter, a transformation matrix from the current sensor coordinate system to the target object coordinate system is determined to obtain a spatial pose transformation relationship; Based on the shared mark of the target image data and the interaction data of each intelligent device, the anchoring type corresponding to the intelligent device is determined; wherein the anchoring type includes single device anchoring and multi-device anchoring; According to the anchoring type, the spatial pose transformation relationship and the local coordinate system corresponding to each intelligent device, the spatial pose anchoring of each intelligent device is realized.

[0008] Optionally, in one or more embodiments of the present application, after the spatial pose of the current frame image data is solved according to the target object coordinate system and the homonym point pair to realize spatial pose anchoring, the method further includes: Obtain historical frame image data to determine historical attitude information corresponding to the historical frame image data; According to the historical attitude information and the current attitude information corresponding to the current frame image data, and the spatial pose transformation relationship, a cumulative transformation relationship is determined; According to the cumulative transformation relationship, the historical frame image data is converted to the target object coordinate system.

[0009] Optionally, in one or more embodiments of the present application, before the spatial pose of the current frame image data is solved according to the target object coordinate system and the homonym point pair to realize spatial pose anchoring, the method further includes: Determine the current environment to load the target image data of the specific target object in the current environment; mapping relationship of pixel coordinates to physical coordinates of the target image data is established to obtain a target object coordinate system of the specific target object according to the mapping relationship.

[0010] Optionally, in one or more embodiments of the present application, after the target image data of the specific target object in the current environment is loaded, the method further comprises: the target image data is denoised, and edge detection is performed on the denoised target image data to obtain a target candidate region of the specific target object; a target region of the specific target object is determined according to a local peak value detection result of the target candidate region; feature points of the target region are extracted to generate a first feature descriptor according to the feature points.

[0011] Optionally, in one or more embodiments of the present application, matching the current frame image data with the target image data to obtain the same name point pair specifically comprises: key frame information corresponding to each current frame image data is obtained, and it is identified whether the current frame image contains the specific target object; wherein the key frame information includes attitude information and a second feature descriptor; If yes, feature matching is performed according to the first feature descriptor and the second feature descriptor to obtain the same name point pair between the target image data and the current frame image data.

[0012] Optionally, in one or more embodiments of the present application, after the current frame image data collected by the sensor is obtained, the method further comprises: monitoring upper application instructions to obtain a target scene corresponding to the upper application instructions; determining whether the current frame image data is subjected to world system transformation based on the target scene and a current scene of the current frame image data; If yes, the attitude information of the current frame image data is transformed according to a transformation relationship corresponding to the world system transformation.

[0013] In a second aspect, one or more embodiments of the present application provide a spatial pose anchoring device, which comprises: a matching unit configured to obtain current frame image data collected by a sensor, and match the current frame image data with target image data to obtain the same name point pair; an anchoring unit configured to solve the spatial pose of the current frame image data according to a target object coordinate system and the same name point pair to realize spatial pose anchoring; wherein the target coordinate system is obtained based on the target image data constructed by the intelligent device in the current environment.

[0014] In a third aspect, one or more embodiments of the present application provide an XR device, comprising: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: execute any of the above-mentioned spatial pose anchoring methods.

[0015] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: By using the same-name point pairs between the current frame image and the target image to solve the pose, positioning can be achieved directly based on the spatial features of the target object, reducing the interference of irrelevant factors in the environment, thereby improving the accuracy of the spatial pose anchoring of the smart device. And through the precise mapping of the target object coordinate system and the same-name point pairs, the stable conversion of the smart device from the local coordinate system to the global coordinate system is achieved, solving the core problem of inconsistent pose references in multi-device collaboration. By constructing the target coordinate system based on the target image data of a specific target object, it is helpful to use the target coordinate system as a global reference system in the future, so that the posture information of the smart device can be converted to this coordinate system for unified expression. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments described in the present application. For those skilled in the art, other drawings can be obtained based on these drawings without inventive work. In the drawings: Figure 1 A schematic diagram of a spatial posture anchoring architecture provided in an embodiment of the present application; Figure 2 A schematic diagram of a flow chart of a spatial posture anchoring method provided in an embodiment of the present application; Figure 3 A schematic diagram of the overall process of a spatial posture anchoring method provided in an embodiment of the present application; Figure 4 A flowchart of a method for achieving spatial posture anchoring of each smart device provided in an embodiment of the present application; Figure 5 A schematic structural diagram of a spatial posture anchoring device provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of an XR device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The embodiments of the present application provide a spatial posture anchoring method, apparatus, and XR device.

[0018] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0019] The embodiments of the present application provide a spatial posture anchoring method, apparatus, and XR device.

[0020] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0021] Unless otherwise defined, technical or scientific terms used in this application shall have the ordinary meanings as understood by persons of ordinary skill in the art to which this application belongs. The terms "a," "an," "the," "these," and similar expressions in this application do not denote limitations on quantity and may be singular or plural. The terms "comprise," "include," "have," and any variations thereof, as used in this application, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include unlisted steps or modules (units) or other steps or modules (units) inherent to the process, method, product, or device. The terms "connected," "connected," "coupled," and similar expressions used in this application are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. As used in this application, "plurality" means two or more. "And / or" describes an association between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone; A and B exist simultaneously; or B exists alone. Generally, the character " / " indicates that the objects in the preceding and following relationship are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0022] The spatial posture anchoring method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, Figure 1As shown, the application environment can include: a smart device terminal 101, a sensor 102 and a specific target object 103. Among them, the smart device terminal 101 can be an augmented reality AR / virtual reality VR head-mounted device, a robot, a drone and other hardware devices that need to align the actual physical scene or need to interact with virtual reality, and can also run in software in the above-mentioned hardware devices. In addition, the smart device terminal can be a single smart device terminal in different scenarios, or a plurality of smart device terminals working cooperatively. The sensor 102 communicates with the smart device terminal 101 and can be used to collect data of the specific target object 103 in the current environment to obtain current frame image data. Then according to the communication between the sensor 102 and the smart device terminal 101, the current frame image data is uploaded to the smart device terminal 101, so that the smart device terminal 101 can match the pre-stored target image data in the current environment with the current frame image data collected by the sensor 102 to obtain a homonym point pair. Among them, the smart sensor terminal 101 is also used to construct a target object coordinate system according to the pre-stored target image data in the current environment. Further, according to the target object coordinate system and the homonym point pair, the spatial pose of the current frame image data is solved to realize spatial pose anchoring. Among them, when a plurality of smart terminal devices terminal 101 anchor based on the same specific target object, the target object coordinate system can be constructed by any one of the plurality of smart device terminals 101, and other smart device terminals 101 can obtain and use the target object coordinate system based on the interaction between the smart device terminals 101.

[0023] As shown, Figure 2 The embodiment of the application provides a flow diagram of a spatial pose anchoring method. Figure 2The method shown can be applied to various scenarios that require high-precision spatial positioning and pose estimation, including but not limited to: augmented reality and virtual reality scenarios, such as in AR glasses or mobile AR applications, by identifying specific target objects such as QR codes, markers, or natural objects, etc., to achieve accurate spatial anchoring of virtual objects, ensuring stable alignment of virtual content with the real world. In the VR environment, based on the target object coordinate system, the real-time pose tracking of headsets, handles and other devices is realized, which improves the immersion and reduces the drift problem. In XR applications, based on the target object coordinate system, a virtual person is placed on a card in the actual scene for interactive use, a building information model (BIM) is placed in the actual construction site to guide construction and acceptance, etc. Robot navigation and autonomous movement scenarios, such as robots identifying pre-set target objects such as landmarks and special patterns to correct their poses in unknown environments, multiple robots sharing the same target object coordinate system to achieve cooperative positioning and task allocation of warehouse logistics robots. Intelligent unmanned industrial automation scenarios, such as AR-assisted maintenance, which can use markers on the device to obtain spatial anchoring of maintenance instructions, corresponding virtual coordinates to actual physical coordinates, to improve operation accuracy. The spatial pose anchoring method is based on the program execution of a smart device, including the following steps: S201-S203.

[0024] S201: Obtain current frame image data collected by a sensor, and match the current frame image data with target image data to obtain a same-named point pair.

[0025] The pose information obtained by the smart device through VIO and other technologies is based on a local coordinate system and cannot be directly associated with a global reference system, i.e., a target object coordinate system, in a physical scene. Therefore, to solve the problem of inconsistent information expression in different reference systems and provide data basis for the transformation from a local system to a global system, in the embodiments of the present application, the current frame image data collected by a camera and other sensors of each smart device, such as XR glasses, intelligent unmanned devices, etc., is obtained. After the current frame image data is preprocessed based on the same target image data, the feature information to be matched is obtained. Through a feature matching algorithm, a same-named point pair is found in the current frame image data and the target image data. It can be understood that the same-named point pair is a pixel point corresponding to the same physical feature point in two images, for example, the pixel coordinates of a corner point on a target object in a pre-stored image and the pixel coordinates of the corner point in the current frame image form a same-named point pair. Since the feature points in the target image data are associated with the target object coordinate system through the mapping from pixel coordinates to physical coordinates, and the feature points of the current frame image correspond to the pixel coordinates of the sensor of the smart device, after matching, a point pair relationship between the 3D coordinates in the target object coordinate system and the 2D pixel coordinates of the current frame image can be established.

[0026] Further, in some embodiments, after acquiring the current frame image data collected by the sensor, the method further comprises: S301: monitoring the upper application instruction to acquire a target scene corresponding to the upper application instruction; S302: determining whether the current frame image data is subjected to world coordinate system transformation based on the target scene and a current scene of the current frame image data; S303: if yes, transforming the attitude information of the current frame image data according to a transformation relationship corresponding to the world coordinate system transformation.

[0027] The intelligent device can run multiple applications simultaneously, and different applications can require different world coordinate systems due to different user needs. Moreover, in some scenarios, when the user or the device switches from one scene to another, such as from an indoor AR game to outdoor navigation, it is necessary to quickly determine whether to switch the reference system, so as to avoid the problem of confusing pose expression. Therefore, to solve the above problems, the intelligent device of the embodiments of the present application will monitor the upper application instruction to acquire the target scene corresponding to the upper application instruction after acquiring the current frame image data collected by each sensor. According to whether the target scene corresponding to the current upper application instruction is consistent with the current scene of the current frame image data, it is determined whether the current frame image data is subjected to world coordinate system transformation. For example, if it is determined based on the upper application instruction that the AR glasses switch from the furniture placement mode to the outdoor navigation mode in a certain application scene, it can be determined that the current scene of the placement mode is based on the room, and the target scene of the outdoor navigation mode is based on the geographical coordinates, which are inconsistent. At this time, the virtual content needs to be automatically re-anchored. Therefore, if transformation occurs, in order to ensure real-time synchronization of the pose data, the attitude information of the current frame image data will be transformed according to the transformation relationship corresponding to the world coordinate system transformation, and the current and historical poses will be corrected according to the transformation relationship to ensure the spatiotemporal continuity.

[0028] This process detects whether the world coordinate system needs to be transformed at present, and when the current frame image data corresponding to all the attitude information needs to be transformed, the target scene is dynamically matched by intelligently monitoring the upper application instruction, thereby realizing flexible switching and seamless connection of the world coordinate system in multiple application scenes, and effectively solving the problem of confusing pose expression caused by the failure of the fixed coordinate system to timely transform the world coordinate system. When the scene change requirement is detected, the world coordinate system transformation is automatically triggered, and the current frame and historical pose data are uniformly corrected, which not only guarantees the spatial consistency of virtual and real content, but also maintains the spatiotemporal continuity across scenes, and provides efficient global pose management capability for multi-task intelligent devices.

[0029] Specifically, in some embodiments, the matching of the current frame image data and the target image data in the above step S201 to obtain the same point pair specifically comprises: S401: Obtain key frame information corresponding to each current frame image data, and identify whether the current frame image contains the specific target object; wherein the key frame information includes: attitude information, a second feature descriptor; S402: If yes, perform feature matching according to the first feature descriptor and the second feature descriptor, and obtain a homonym point pair between the target image data and the current frame image data.

[0030] In order to convert the local coordinate system of the current frame image data into the target object coordinate system, while avoiding the problem of large amount of calculation and easy interference caused by directly searching for the target object in the full image in the traditional way. In the embodiment of the present application, as shown in Figure 3 the key frame information includes: attitude information, i.e. VIO pose, feature points and a second feature descriptor. Then, if it is determined based on the above judgment that the current frame image contains the specific target object, feature matching is performed according to the first feature descriptor and the second feature descriptor, and a homonym point pair between the target image data and the current frame image data is obtained. For example, if the descriptor is a binary type descriptor, Hamming distance matching can be used, and for a floating point type descriptor, nearest neighbor ratio test can be used, so as to retain point pairs with matching scores higher than a threshold value, and obtain the homonym point pair.

[0031] S202: According to the target object coordinate system and the homonym point pair, the spatial pose of the current frame image data is solved, and spatial pose anchoring is realized; wherein the target coordinate system is obtained by constructing the target image data based on the intelligent device in the current environment. The intelligent device for constructing the target coordinate system can be the execution subject of the present application, or can be constructed by other intelligent devices in advance. It can be understood that the method of the present application can be that the execution subject constructs the target coordinate system, and then enters the virtual space corresponding to the target coordinate system through the target image data again, or the execution subject enters the virtual space corresponding to the target coordinate system constructed by other terminal devices through the target image data, so as to realize the synchronization of the virtual spaces between multiple devices.

[0032] The pose obtained by the current intelligent device through VIO or SLAM is usually based on a local coordinate system, which cannot directly express the position of the target object in the real physical world, resulting in that the virtual content cannot be accurately superimposed or the devices cannot be cooperatively positioned. Moreover, the traditional method directly uses 2D-2D matching to solve the pose, which is easily affected by the false matching points, resulting in pose jitter or drift problems, and the observation poses of different devices on the same target object cannot be directly aligned due to the difference in the local coordinate system, making it difficult to realize interaction. Therefore, in order to solve the above problems, after obtaining the same point pair based on the above step S201, through the 3D-2D correspondence relationship between the target object coordinate system and the same point pair, the 3D-2D data association of the same point can be realized, and the correct data association can be obtained to solve the relative pose. That is, in the embodiment of the present application, according to the target coordinate system and the same point pair, the spatial pose of the current frame image data is solved, the relative pose of the current frame image data and the target image data is obtained, the target object pose output or the transformation of the reference frame is realized, and the spatial pose anchoring of each intelligent device is completed. Through the accurate mapping of the target object coordinate system and the same point pair, the stable conversion of the intelligent device from the local coordinate system to the global coordinate system is realized, and the core problem of the non-uniformity of the pose reference in multi-device cooperation is solved. It should be noted that the target object coordinate system is obtained by the current intelligent device or other intelligent devices in the current environment, and is constructed in advance according to the target image data.

[0033] The target object coordinate system constructed here is because when the existing intelligent device performs spatial pose anchoring, the carrier state information obtained by the intelligent device through SLAM and other technologies is usually based on a local coordinate system, and such a local coordinate system is difficult to directly express the pose of a specific object in a global scene, resulting in that the information of the intelligent device required by the user is difficult to be uniformly managed and interacted. Moreover, when the device positioning information is lost, the local coordinate system needs to be aligned to the global reference system, but at present, especially in the scene of multi-intelligent device cooperation, there is a lack of accurate way to align to the global reference system. Therefore, in order to unify the global coordinate system and provide a unified reference for the cooperation of multiple intelligent devices, in the embodiment of the present application, the target image data of a specific target object is loaded through the specified path of the intelligent device, and the target coordinate system is constructed based on the target image data. By constructing the target coordinate system based on the target image data, the target coordinate system can be used as a global reference system in the subsequent process, so that the pose information of the intelligent device can be converted to the coordinate system, realizing unified expression. The specific target object can be set based on the user's demand in different scenes, such as a two-dimensional code, a poster, etc.

[0034] In some feasible embodiments, the target coordinate system can be a right-hand global coordinate system with the x-axis pointing to the right and the y-axis pointing downward, which is fixedly connected with the specific target object. The origin of the target object coordinate system is usually set at the center of the target object, and the axis direction is consistent with the actual scene direction, so as to facilitate the subsequent pose transformation to directly reflect the relative relationship between the device and the physical scene.

[0035] The intelligent device can convert the specific target object into a global reference system by obtaining the target image data of the specific target object and then constructing the target object coordinate system corresponding to the specific target object according to the target image data, thereby solving the problems of local coordinate system unable to be globally expressed, difficulty in multi-device cooperation, and lack of virtual-real interaction reference, etc., providing a unified coordinate framework for subsequent pose anchoring, and providing a stable and reliable foundation for realizing precise spatial positioning and interaction.

[0036] In some embodiments, before the step S202 of solving the spatial pose of the current frame image data according to the target object coordinate system and the same-named point pair, the method further includes: S501: determining a current environment to load target image data of a specific target object in the current environment; S502: establishing a mapping relationship between pixel coordinates and physical coordinates of the target image data, so as to obtain a target object coordinate system fixedly connected with the specific target object according to the mapping relationship.

[0037] The current specific target object is a self-defined marker for providing a reference for spatial pose anchoring of the intelligent device. For example, if a virtual object needs to be accurately superimposed on a specific position of a real scene, such as a card, in an AR / VR device, in order to provide an anchor point for virtual-real fusion and ensure alignment of virtual content and the physical world, the current environment of the intelligent device is determined in the embodiments of the present application, so as to load the target image data corresponding to the specific target object preexisting in the database in the current environment. For example, if a virtual object needs to be accurately superimposed on a specific position of a real scene, such as a card, in an AR / VR device, the specific target object in this scene can be a two-dimensional code or a bar code and other self-defined patterns. Then, a mapping relationship between pixel coordinates and physical coordinates of the target image data is established, so as to obtain a target object coordinate system fixedly connected with the specific target object according to the mapping relationship. The mapping relationship between pixel coordinates and physical coordinates of the target image data can be as follows: ; ; .

[0038] Wherein, u, v are the detected feature pixel coordinates, pixelcols, pixelrows are the pixel width and pixel height of the target image, markerWidth, markerHeight are the actual physical width and actual physical height of the target image data measurement, x, y are the coordinates in the converted target object coordinate system.

[0039] Through the explicit mapping relationship, the target image data of the specific target object is associated with the actual physical data, thereby constructing a global coordinate system strictly fixed with the specific target object. This mapping method ensures the accuracy of pose calculation, so that the intelligent device can realize stable spatial pose anchoring based on a unified physical reference system, effectively solving the problem of aligning the local coordinate system with the global scene, while supporting the high-precision requirements of multi-device collaboration and virtual-real interaction. In addition, the mapping process given by the above process directly based on the pixel width and height of the target image, the actual physical width and height, establishes an intuitive correspondence between the pixel coordinates (u, v) and the target object coordinate system coordinates (x, y), avoiding complex spatial geometric calculations, reducing the difficulty and cost of coordinate conversion, and facilitating efficient implementation in intelligent devices.

[0040] Wherein, in a feasible embodiment, the current environment of the intelligent device is determined to load the target image data of the specific target object in the current environment: since different intelligent devices may be deployed in different environments and are affected by different light conditions, for example, the current environment data of each intelligent device can be obtained, so as to match the current environment data with the environment adaptation range corresponding to each specific target object in the pre-set database, thereby determining the specific target object through dynamic environment adaptation, improving the robustness and practicality of detection, so that the spatial pose anchoring method can be reliably and accurately implemented in different light, distance and occlusion scenes.

[0041] Further, in some embodiments, after the step S301 of loading the target image data of the specific target object in the current environment, the method further comprises: S601: performing noise reduction processing on the target image data, and performing edge detection on the noise-reduced target image data to obtain a target candidate region of the specific target object; S602: determining a target region of the specific target object according to a local peak value detection result of the target candidate region; S603: extracting feature points of the target region to generate a first feature descriptor according to the feature points.

[0042] The target image data pre-stored in the storage unit of the intelligent device can have interference such as shadow, dirt or background noise interference, so that the image noise of the target image data can be eliminated through noise reduction processing to improve the signal-to-noise ratio of the specific target object. Then, the Canny or Sobel operator is used to detect the edge of the target image data after noise reduction, so as to generate a target candidate area through connected domain analysis. The gradient amplitude of the target image data in each target candidate area is calculated to obtain a local peak detection result, and the local maximum point is retained to determine the target area of the specific target object. For example, in a poster, irrelevant text patterns are excluded, and only the LOGO area is retained as the target area corresponding to the specific target object. Then, the feature points of the target image data in the target area are extracted, and a first feature descriptor is generated according to the feature points, which provides high-discrimination data for subsequent 3D-2D matching and ensures the pose solving accuracy. This process determines the target area corresponding to the specific target object, and generates the first feature descriptor based only on the target area, thereby reducing the interference of non-target areas. In addition, it can also be understood that: since the feature points have invariance characteristics, it can be ensured that the features of the pre-stored target image data and the current frame image data collected by the intelligent device in real time can still be effectively matched under different shooting angles and distances, which helps to provide reliable corresponding relationship for subsequent 3D-2D matching.

[0043] In some possible embodiments, according to the local peak detection result of the target candidate area, the target area of the specific target object can be determined according to the local peak detection result combined with the gradient amplitude and phase of the target candidate area, specifically: In a complex scene, the target candidate region determined by the edge detection process of the Canny or Sobel operator may be misjudged due to noise or texture interference. Therefore, in order to accurately distinguish the real target edge from the noise by combining the gradient amplitude reflecting the pixel change and the gradient phase reflecting the pixel change, and improve the reliability of target region positioning, the target candidate region image is convolved by the Sobel or Scharr operator in the embodiment of the application to obtain the horizontal gradient and the vertical gradient. Then, the gradient amplitude corresponding to each target candidate region image is obtained according to the square root of the sum of the squares of the horizontal gradient and the vertical gradient. The gradient phase corresponding to each target candidate region image is obtained based on the arctangent of the ratio of the vertical gradient to the horizontal gradient. The gradient amplitude is used to represent the edge strength, and the greater the value, the more likely it is a target boundary; the gradient phase is used to represent the edge direction, such as 0° for a vertical edge and 90° for a horizontal edge. According to the obtained gradient amplitude and gradient phase, the local peak value of the target candidate region is detected to obtain a local peak value detection result, so as to determine the target region of the specific target object according to the local peak value detection result. This method upgrades the single-dimensional judgment to multi-dimensional collaborative verification through joint analysis of the gradient amplitude and phase, significantly improving the accuracy and robustness of target region detection.

[0044] In some embodiments, the spatial pose of the current frame image data is solved according to the target object coordinate system and the homonym point pair, and the spatial pose of each intelligent device is anchored, specifically including: S701: According to the construction rule of the target object coordinate system, the current frame image data in the homonym point is mapped to obtain the matching relationship of the homonym point pair; S702: Based on the matching relationship of the homonym point pair, the spatial pose transformation relationship of the current frame image data relative to the target object coordinate system is obtained, and the spatial pose of each intelligent device is anchored.

[0045] Firstly, based on the pre-constructed target object coordinate system, such as a right-handed coordinate system with the center of the two-dimensional code as the origin, the X-axis to the right and the Y-axis to the down, the detected 2D image feature points are mapped to the 3D physical space through the perspective projection relationship, the accurate conversion from pixel coordinates (u, v) to physical coordinates (x, y, z) is established, and it is ensured that each matching point pair has real physical coordinates. Then, according to the matching relationship of the homonym point pair, the pose of the intelligent device relative to the target object coordinate system is solved from the matching relationship, and then the spatial pose transformation relationship of the current frame image data relative to the target object coordinate system is determined, and the conversion from the local coordinate system to the global coordinate system is completed to anchor the spatial pose of each intelligent device. Compared with the traditional SLAM positioning method, the absolute measurement reference is provided by the target object coordinate system of the specific target object, and the cumulative error is eliminated.

[0046] In the application scenario of an embodiment, when AR glasses are needed, the device parameter virtual label is accurately superimposed on the transformer nameplate in reality. At this time, the two-dimensional code on the nameplate can be selected as a specific target object to construct a right-hand coordinate system: the origin is the center of the two-dimensional code; the X-axis is parallel to the long side of the two-dimensional code to the right; the Y-axis is parallel to the short side of the two-dimensional code downward; and the Z-axis is perpendicular to the surface of the nameplate outward. Then, the AR glasses camera captures the current frame image data to detect the two-dimensional code, extracts the key frame data to construct a feature descriptor, matches the feature descriptor of the current frame image data with the feature descriptor of the target image data, and obtains corresponding point pairs. Assuming that the corresponding point pairs are four corner points, at this time, the pose transformation can be calculated based on the existing pose transformation algorithm, and the pose transformation is converted into a transformation matrix, and then the virtual device label is converted from the local coordinate system to the target object coordinate system according to the transformation matrix, so as to realize the spatial pose anchoring of the smart device.

[0047] Further, as Figure 4 In some embodiments, the step S702 obtains a spatial pose transformation relationship of the current frame image data relative to the target object coordinate system based on the matching relationship of the corresponding point pairs, and realizes the spatial pose anchoring of each smart device, specifically including: S801: determining a transformation matrix from the current sensor coordinate system to the target object coordinate system according to the matching relationship and the sensor intrinsic parameter, and obtaining a spatial pose transformation relationship; S802: determining an anchoring type corresponding to each smart device based on the shared mark of the target image data and the interaction data of each smart device; wherein the anchoring type includes single-device anchoring and multi-device anchoring; S803: realizing the spatial pose anchoring of each smart device according to the anchoring type, the spatial pose transformation relationship, and the local coordinate system corresponding to each smart device.

[0048] Based on the mapping relationship between the pixel coordinates of the target image data and physical coordinates determined in steps S301-S303, the feature points of the target image data in the same-name point pair have been converted to 3D coordinates in the target coordinate system through the pixel-to-physical coordinate mapping relationship. The feature points of the current frame image correspond to the 2D pixel coordinates of the smart device sensor. Then, by combining the sensor's intrinsic parameters and using existing 3D-2D pose resolution methods, the transformation matrix of the current sensor coordinate system relative to the target coordinate system, i.e., the spatial pose transformation relationship, can be calculated. Based on the shared tags of the target image data and the interaction data between devices, the anchoring type corresponding to the smart device is determined to be single-device anchoring or multi-device anchoring. It can be understood that single-device anchoring only requires converting the pose information of a single smart device to the target coordinate system, such as when a single pair of XR glasses overlays a virtual person on a real-world card. Multi-device anchoring requires unifying the pose information of multiple smart devices into the same target coordinate system, such as when multiple smart unmanned devices collaborate on a construction site and are anchored based on the same BIM model. Then, according to the determined anchoring type, spatial posture transformation relationship and the local coordinate system corresponding to each smart device, the spatial posture anchoring of each smart device is realized.

[0049] The current posture information obtained by smart devices through technologies such as VIO is based on the local coordinate system and cannot be directly associated with the global reference system of the physical scene. This process solves the problem of accurate conversion from the local system to the global system by determining the transformation matrix from the current sensor coordinate system to the target object coordinate system, so that the device posture can be expressed in a unified reference system. By distinguishing the anchoring types, it also solves the problem that a single anchoring logic cannot adapt to multiple scenes, ensuring the precise positioning of a single device and the collaborative consistency of multiple devices. In addition, the local coordinate systems of multiple smart devices are independent and may have cumulative errors. If directly associated, it will lead to collaborative deviation. However, this process solves the problem of inconsistent posture reference systems of multiple devices through unified conversion of the target object coordinate system, ensuring the spatial coordination of interactions between devices.

[0050] In one feasible embodiment, determining the anchor type corresponding to the smart device based on the shared mark of the target image data and the interaction data of each smart device specifically includes: The specific target object is acquired as a corresponding shared marker to determine a shared device range corresponding to the specific target object according to the shared marker. Interaction data of each smart device is acquired to determine, according to the interaction data, that there is an associated smart device of a collaborative task in the current scene. If the associated smart device exists in the shared device range, it is determined that the anchor type corresponding to the smart device is a multi-device anchor. If it is determined that the associated device does not belong to the shared device range, the shared permission corresponding to the associated device is determined based on the data type of the interaction data and the target object data corresponding to the historical anchor data of the associated device. If the shared permission is greater than the preset permission, the shared marker is opened to the associated device at this time, and it is determined that the anchor type corresponding to the smart device is a multi-device type. Otherwise, if the shared permission is less than the preset permission, the anchor type corresponding to the smart device is a single-device type.

[0051] In the implementation of the feasible embodiment, based on the physical range limitation of the shared marker and the real-time analysis of the interaction data, the system can intelligently identify an effective collaborative device group and ensure accurate synchronization of pose data. Moreover, the permission grading strategy realizes fast authorization of temporary devices under the premise of ensuring system security through historical anchor data learning and data type verification, so that the anchor process can maintain high-precision collaboration of the core device group and dynamically expand temporary operation devices, facilitating multiple smart devices to be anchored to the same target coordinate system through the same specific target object for interaction.

[0052] In a feasible embodiment, in the process of realizing the spatial pose anchoring of each smart device according to the anchor type, the spatial pose transformation relationship and the local coordinate system corresponding to each smart device, for single-device anchoring, the pose information of the smart device in the local coordinate system can be directly converted to the target object coordinate system based on the transformation matrix obtained in step S801. For multi-device anchoring, the local coordinate system pose of each device is first converted to the target object coordinate system through the spatial pose transformation relationship between each device and the target object, and then the deviation is calibrated based on the interaction data between devices to ensure that the poses of all devices are consistent in the global reference system, and finally the spatial pose anchoring of multiple devices is realized. In this process, by distinguishing between single-device and multi-device anchor types, the technical solution can be adapted to different application scenarios, improving the universality and practicality of the solution. After converting the pose information of all devices to the target object coordinate system, the problem of information not being able to be expressed globally in the local coordinate system is solved, and the unified target object coordinate system enables virtual content to be stably superimposed on the real scene, and multiple devices can realize precise collaboration based on the same reference system, improving the application value of smart devices in XR, industrial collaboration and other scenarios.

[0053] Further, in some embodiments, after the spatial pose solving and spatial pose anchoring of the current frame image data according to the target object coordinate system and the homonym point pair, the method further comprises: S901: obtaining historical frame image data to determine historical pose information corresponding to the historical frame image data; S902: determining a cumulative transformation relationship according to the historical pose information, current pose information corresponding to the current frame image data, and the spatial pose transformation relationship; S903: converting the historical frame image data to the target object coordinate system according to the cumulative transformation relationship.

[0054] During the movement of the intelligent device, the VIO or SLAM system of the intelligent device continuously outputs pose information in a local coordinate system. After the target object is detected and the pose anchoring of the current frame is completed, the historical frame data still stays in the original local coordinate system, which leads to inconsistent expression of the poses of the previous and current frames and makes it impossible to analyze the continuous trajectory or perform global optimization. In addition, the historical pose in the local coordinate system may also accumulate errors due to sensor drift. Therefore, after obtaining the historical frame image data of the intelligent device based on step S901 in the embodiments of the present application to obtain historical pose information corresponding to the historical frame image data, a cumulative transformation relationship is determined according to the historical pose information, current pose information corresponding to the current frame image data, and the spatial pose transformation relationship. It can be understood that the historical pose information refers to historical frame pose data output by the VIO or SLAM system before the target object is detected, based on the original local coordinate system; the current pose information refers to the pose data of the current frame in the original local coordinate system after the target object is detected; and the spatial pose transformation relationship is a transformation matrix of the current frame coordinate system to the target object coordinate system obtained through the homonym point pair matching and pose solving. Since the historical pose information and the current pose information are both based on the same original local coordinate system, the relative transformation matrix of the historical frame to the current frame can be calculated through the pose difference between the historical pose information and the current pose information, and then the cumulative transformation relationship can be obtained through the relative transformation matrix and the spatial pose transformation relationship. All historical poses are converted to the target object coordinate system through the cumulative transformation relationship, thereby forming continuous image data without jumps.

[0055] In the process, by determining the cumulative transformation relationship and uniformly converting the historical poses to the target object coordinate system, the problem of split of the pose reference system of the previous and subsequent frames caused by the historical frames staying in the original local coordinate system and the current frame being anchored to the target object coordinate system is solved, so that the pose information of all frames is unified in the target object coordinate system. In the virtual-real fusion scene such as XR, the virtual object can maintain a stable superposition relationship with the real target object as the device moves or time elapses, and the shaking or mispositioning of the virtual content caused by the jump of the coordinate system is avoided. Moreover, the historical poses in the local coordinate system are prone to error accumulation due to the sensor drift of the VIO or SLAM system, while the target object coordinate system is based on the fixed reference of the physical world. However, after the historical poses are converted to the coordinate system through the cumulative transformation, the stability of the global reference system can be used to constrain and correct the errors, thereby improving the reliability of the long-term pose data.

[0056] The embodiments of the present application also provide a spatial pose anchoring device of an intelligent device, as shown in Figure 5 The spatial pose anchoring device of the intelligent device comprises: A matching unit 501 is configured to acquire current frame image data collected by a sensor, match the current frame image data with target image data, and obtain corresponding point pairs. An anchoring unit 502 is configured to solve the spatial pose of the current frame image data according to a target object coordinate system and the corresponding point pairs, and realize spatial pose anchoring. The target coordinate system is obtained by constructing the target image data by the intelligent device in the current environment.

[0057] In some embodiments, solving the spatial pose of the current frame image data according to the target object coordinate system and the corresponding point pairs, and realizing spatial pose anchoring, specifically comprises: Mapping the current frame image data in the corresponding point pairs according to the construction rule of the target object coordinate system, to obtain the matching relationship of the corresponding point pairs; Obtaining the spatial pose transformation relationship of the current frame image data relative to the target object coordinate system based on the matching relationship of the corresponding point pairs, to perform spatial pose anchoring based on the spatial pose transformation relationship.

[0058] In some embodiments, obtaining the spatial pose transformation relationship of the current frame image data relative to the target object coordinate system based on the matching relationship of the corresponding point pairs, to perform spatial pose anchoring based on the spatial pose transformation relationship, specifically comprises: Determining the transformation matrix from the current sensor coordinate system to the target object coordinate system according to the matching relationship and the sensor intrinsic parameters, to obtain the spatial pose transformation relationship; determine an anchor type corresponding to the smart device based on the shared mark of the target image data and the interaction data of each smart device, wherein the anchor type comprises a single-device anchor and a multi-device anchor; implement spatial pose anchoring of each smart device according to the anchor type, the spatial pose transformation relationship, and the local coordinate system corresponding to each smart device.

[0059] In some embodiments, after the spatial pose anchoring is implemented by solving the spatial pose of the current frame image data according to the target object coordinate system and the homonym point pair, the method further comprises: obtain historical frame image data to determine historical pose information corresponding to the historical frame image data; determine a cumulative transformation relationship according to the historical pose information, the current pose information corresponding to the current frame image data, and the spatial pose transformation relationship; convert the historical frame image data to the target object coordinate system according to the cumulative transformation relationship.

[0060] In some embodiments, before the spatial pose anchoring is implemented by solving the spatial pose of the current frame image data according to the target object coordinate system and the homonym point pair, the method further comprises: determine a current environment to load target image data of a specific target object in the current environment; establish a mapping relationship between pixel coordinates of the target image data and physical coordinates to obtain a target object coordinate system fixedly connected to the specific target object according to the mapping relationship.

[0061] In some embodiments, after the target image data of the specific target object in the current environment is loaded, the method further comprises: perform noise reduction processing on the target image data, and perform edge detection on the noise-reduced target image data to obtain a target candidate region of the specific target object; determine a target region of the specific target object according to a local peak detection result of the target candidate region; extract feature points of the target region to generate a first feature descriptor according to the feature points.

[0062] In some embodiments, matching the current frame image data and the target image data to obtain a homonym point pair specifically comprises: obtain key frame information corresponding to each current frame image data, and identify whether the current frame image contains the specific target object; wherein the key frame information includes pose information and a second feature descriptor; If yes, feature matching is performed according to the first feature descriptor and the second feature descriptor to obtain corresponding point pairs between the target image data and the current frame image data.

[0063] In some embodiments, after the current frame image data collected by the sensor is acquired, the method further includes: Monitoring an upper application instruction to acquire a target scene corresponding to the upper application instruction; Determining whether the current frame image data is subjected to world system transformation based on the target scene and a current scene of the current frame image data; If yes, transforming pose information of the current frame image data according to a transformation relationship corresponding to the world system transformation.

[0064] Embodiments of the present application also provide an XR device, which includes: At least one processor; and, A memory in communication connection with the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any of the above-mentioned spatial pose anchoring methods.

[0065] In some embodiments of the present application, as shown in Figure 6 The XR device can include a computing unit 601, a ROM 102, a RAM 103, a bus 604, an I / O interface 605, a front-end VIO 106, a sensor input unit 607, a sensor input unit 607, an output unit 608, a storage unit 609, and a communication unit 610.

[0066] The computing unit 601 can execute the spatial pose anchoring processing and analysis process according to the computer program stored in the read-only memory 602 or the computer program loaded from the storage unit 610 to the random access memory RAM 603. In the RAM 603, various programs and data required for XR device operation can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other through the bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0067] The plurality of components in the XR device are connected to the I / O interface 605, including: a front-end VIO 606 to which an input unit 607, which can be various sensors, is connected; an output unit 608, such as various types of displays, speakers, etc.; a storage unit 609, such as a magnetic disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 610 allows the smart device to interact with other XR devices in information or data through a computer network such as the Internet and / or various telecommunications networks.

[0068] The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence computing chips, various computing units running machine learning model algorithms, a digital signal processor, and any appropriate processor, controller, microcontroller, etc. In the process of implementing the spatial pose anchoring method, the computing unit 601 obtains target image data of a specific target object to construct a target object coordinate system corresponding to the specific target object; obtains current frame image data collected by a sensor, and matches the current frame image data with the target image data to obtain homonymous point pairs; and according to the target object coordinate system and the homonymous point pairs, performs spatial pose solving on the current frame image data to realize various processing processes of spatial pose anchoring. For example, in some embodiments, the spatial pose anchoring method can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 609. In some embodiments, part or all of the computer program can be loaded and / or installed on the XR device via the ROM 602 and / or the communication unit 610. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps in the embodiments described below can be performed. In other possible embodiments, the computing unit 601 can be configured to perform the spatial pose anchoring method of the present application by any other appropriate means.

[0069] Each of the embodiments in the present application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0070] The above-described embodiments of the application have several aspects, no single one of which is solely responsible for the application's desirable attributes. Without limiting the scope of this application, other aspects of the application will become apparent from consideration of the drawings and following detailed description, it being understood that such changes in the state of the art can be made without departing from the spirit and scope of the application.

[0071] The specific embodiments described herein have many advantages resulting from the particular nature of the application. However, various modifications and changes can be made to such specific embodiments without departing from the overall spirit of the application. For example, the described embodiments can be modified to operate in different environments. It should be understood that the described embodiments are only a few of numerous specific embodiments that can be realized. It is therefore desired that the application be considered in its broadest form, including the following claims, to enable just the scope of the application.

Claims

1. A spatial pose anchoring method, characterized in that: The method comprises: Acquire the current frame image data collected by the sensor, and match the current frame image data with the target image data to obtain the same-name point pairs; According to the target object coordinate system and the pair of points with the same name, the spatial pose of the current frame image data is solved to achieve spatial pose anchoring; wherein, the target coordinate system is obtained by constructing the target image data based on the smart device in the current environment.

2. A spatial pose anchoring method according to claim 1, characterized in that: Solving the spatial pose of the current frame image data according to the target object coordinate system and the pair of points with the same name to achieve spatial pose anchoring specifically includes: Mapping the current frame image data within the same-name points according to the construction rules of the target object coordinate system to obtain a matching relationship between the same-name point pairs; Based on the matching relationship of the pairs of points with the same name, a spatial pose transformation relationship of the current frame image data relative to the target object coordinate system is obtained, so as to perform spatial pose anchoring based on the spatial pose transformation relationship.

3. A spatial pose anchoring method according to claim 2, characterized in that: Based on the matching relationship of the pairs of points with the same name, obtaining the spatial pose transformation relationship of the current frame image data relative to the target object coordinate system, so as to perform spatial pose anchoring based on the spatial pose transformation relationship, specifically including: Determine the transformation matrix from the current sensor coordinate system to the target object coordinate system based on the matching relationship and the sensor internal parameters to obtain the spatial pose transformation relationship; Determine the anchoring type corresponding to the smart device based on the shared mark of the target image data and the interaction data of each smart device; wherein the anchoring type includes: single-device anchoring and multi-device anchoring; The spatial posture anchoring of each smart device is achieved according to the anchoring type, the spatial posture transformation relationship and the local coordinate system corresponding to each smart device.

4. A spatial pose anchoring method according to claim 1, characterized in that: After performing spatial pose solving on the current frame image data based on the target object coordinate system and the pair of points of the same name to achieve spatial pose anchoring, the method further includes: Acquiring historical frame image data to determine historical posture information corresponding to the historical frame image data; Determining a cumulative transformation relationship based on the historical posture information, the current posture information corresponding to the current frame image data, and the spatial posture transformation relationship; According to the accumulated transformation relationship, the historical frame image data is transformed into a target object coordinate system.

5. A spatial pose anchoring method according to claim 1, characterized in that: Before performing spatial pose solving on the current frame image data based on the target object coordinate system and the pair of points of the same name to achieve spatial pose anchoring, the method further includes: Determining a current environment to load target image data of a specific target object in the current environment; A mapping relationship between pixel coordinates of the target image data and physical coordinates is established to obtain a target coordinate system fixed to the specific target according to the mapping relationship.

6. A spatial pose anchoring method according to claim 5, characterized in that: After loading the target image data of the specific target object in the current environment, the method further includes: Performing noise reduction processing on the target image data, and performing edge detection on the noise-reduced target image data to obtain a target candidate area of ​​the specific target object; Determining a target area of ​​the specific target object according to a local peak detection result of the target candidate area; Feature points of the target area are extracted to generate a first feature descriptor according to the feature points.

7. A spatial pose anchoring method according to claim 6, characterized in that: Matching the current frame image data with the target image data to obtain pairs of points with the same name, specifically including: Obtaining key frame information corresponding to each current frame image data, and identifying whether the current frame image contains the specific target object; wherein the key frame information includes: posture information and a second feature descriptor; If so, feature matching is performed based on the first feature descriptor and the second feature descriptor to obtain the same-name point pairs between the target image data and the current frame image data.

8. A spatial pose anchoring method according to claim 1, characterized in that: After acquiring the current frame image data collected by the sensor, the method further includes: Monitoring upper-layer application instructions to obtain target scenarios corresponding to the upper-layer application instructions; determining whether the current frame image data is to be subjected to a world system transformation based on the target scene and the current scene of the current frame image data; If so, the posture information of the current frame image data is transformed according to the transformation relationship corresponding to the world system transformation.

9. A spatial posture anchoring device, characterized in that: The device comprises: a matching unit, configured to obtain current frame image data collected by the sensor, and match the current frame image data with target image data to obtain pairs of points with the same name; An anchoring unit is used to solve the spatial pose of the current frame image data according to the target object coordinate system and the pair of points with the same name, so as to realize spatial pose anchoring; wherein, the target coordinate system is obtained by constructing the target image data based on the intelligent device in the current environment.

10. An XR device, characterized in that: The XR device includes: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: execute the spatial pose anchoring method described in any one of claims 1-8 above.

Images