Object pose estimation method, device, equipment and medium

Abstract

The invention relates to an object pose estimation method and device, equipment and a medium. The object pose estimation method comprises the following steps: acquiring three-dimensional point cloud data of a plurality of objects; performing registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information; in the registration processing process, after registration information corresponding to the successfully registered target three-dimensional point cloud data is obtained, the successfully registered target three-dimensional point cloud data is removed from the three-dimensional point cloud data, and registration processing continues until a registration stopping condition is met; and performing pose conversion on the registration information to obtain pose information of the object corresponding to the registration information. According to the embodiment of the invention, the method can avoid the impact on the registration information from a complex illumination environment, and improves the accuracy of the pose information of a plurality of objects.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method, device, equipment and medium for estimating the position and posture of an object. Background Art

[0002] At present, before completing the task of grasping the target object, it is necessary to obtain the position and posture information of the target object, and then to carry out subsequent processing work. At present, most of the two-dimensional images of the target object are collected, and then the position and posture information of the target object is obtained by using the two-dimensional image. However, in some special scenes (such as in a factory environment, there are multiple target objects), due to the complex lighting environment and the possible stacking of the target objects, the characteristics of the target objects in the two-dimensional image vary greatly, resulting in missed detection and false detection of the target objects, which also affects the accuracy of the target object pose estimation.

[0003] Therefore, how to more accurately obtain the position and posture information of the target object in these special scenes becomes an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a method, device, equipment and medium for object pose estimation.

[0005] In a first aspect, the present disclosure provides a method for estimating an object pose, comprising:

[0006] Obtain 3D point cloud data of multiple objects;

[0007] Perform registration processing on the three-dimensional point cloud data according to the preset registration template to obtain corresponding registration information;

[0008] During the registration process, after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data, the successfully registered target three-dimensional point cloud data is removed from the three-dimensional point cloud data, and the registration process is continued until the registration stop condition is reached;

[0009] The registration information is transformed into a posture to obtain the posture information of the object corresponding to the registration information.

[0010] In a second aspect, the present disclosure provides an object pose estimation device, comprising:

[0011] A data acquisition module, used to obtain three-dimensional point cloud data of multiple objects;

[0012] A registration processing module is used to perform registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information;

[0013] The object removal module is used to remove the successfully registered target three-dimensional point cloud data from the three-dimensional point cloud data after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data during the registration process, and continue the registration process until the registration stop condition is reached;

[0014] The posture conversion module is used to convert the registration information into posture information to obtain the posture information of the object corresponding to the registration information.

[0015] In a third aspect, the present disclosure provides an object pose estimation device, comprising:

[0016] processor;

[0017] A memory for storing executable instructions;

[0018] The processor is used to read executable instructions from the memory and execute the executable instructions to implement the object pose estimation method of the first aspect.

[0019] In a fourth aspect, the present disclosure provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor implements the object pose estimation method of the first aspect.

[0020] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0021] The object pose estimation method, device, equipment and medium of the disclosed embodiment can obtain the three-dimensional point cloud data of multiple objects, and then perform registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information. Then, during the registration process, after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data, the successfully registered target three-dimensional point cloud data is removed from the three-dimensional point cloud data, and the registration process is continued until the registration stop condition is reached. Finally, the registration information is converted into pose to obtain the pose information of the object corresponding to the registration information. In this way, only the three-dimensional point cloud data of the object is collected, and the registration process is performed according to the three-dimensional point cloud data. Moreover, the successfully registered target object can be removed to ensure that all the multiple objects have completed the registration process, thereby avoiding the influence of complex lighting environment on the registration information and improving the accuracy of the pose information of multiple objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0023] Figure 1A flowchart of an object pose estimation method provided by an embodiment of the present disclosure;

[0024] Figure 2 A schematic diagram of three-dimensional point cloud data provided by an embodiment of the present disclosure;

[0025] Figure 3 A schematic diagram of a successfully registered three-dimensional point cloud data provided by an embodiment of the present disclosure;

[0026] Figure 4 A schematic diagram of a target object removal result provided by an embodiment of the present disclosure;

[0027] Figure 5 A schematic diagram of another target object removal result provided by an embodiment of the present disclosure;

[0028] Figure 6 A schematic diagram of posture information provided by an embodiment of the present disclosure;

[0029] Figure 7 A schematic diagram of the structure of an object posture estimation device provided in an embodiment of the present disclosure;

[0030] Figure 8 A schematic diagram of the structure of an object pose estimation device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0031] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0032] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0033] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0034] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0035] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0036] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0037] In the workshop scenario, every time a material frame containing parts arrives at the area to be grasped, if a robotic arm is required to grasp the parts, the specific pose information of the parts in the material frame must be obtained. Therefore, a visual positioning system can be introduced to estimate the pose of the target parts in each material frame. Since there are more than one parts in the material frame, the pose of each target part needs to be estimated. Since the lighting environment in the workshop is relatively complex and there is stacking between parts, the features of the two-dimensional (2D) image vary greatly, and the resulting 2D ROI may suffer from missed detections and false detections, affecting the accuracy of pose estimation.

[0038] In order to solve the above problems, the present disclosure provides a method, device, equipment and medium for estimating the position and posture of an object. Figures 1 to 6 The object pose estimation method provided by the embodiment of the present disclosure is described in detail.

[0039] Figure 1 A flow chart of an object pose estimation method provided by an embodiment of the present disclosure is shown.

[0040] In the embodiment of the present disclosure, the object pose estimation method may be performed by an electronic device. Specifically, the electronic device may include but is not limited to a mobile terminal such as a computer device, a mobile phone, a vehicle-mounted device, a vehicle controller, a tablet computer, a wearable device, etc.

[0041] like Figure 1 As shown, the object pose estimation method may include the following steps.

[0042] S110, obtaining three-dimensional point cloud data of multiple objects.

[0043] In an embodiment of the present disclosure, an electronic device may acquire three-dimensional point cloud data of multiple objects.

[0044] Optionally, the three-dimensional point cloud data may be point cloud data of the object in a spatial coordinate system, wherein the point cloud data may be a collection of coordinate data of multiple points in the spatial coordinate system.

[0045] Specifically, the electronic device can obtain three-dimensional point cloud data of multiple objects collected by devices such as cameras and laser scanners. For example, in a workshop scene, a material frame containing parts may include multiple parts, that is, multiple objects. The electronic device can collect multiple parts in the material frame through devices such as cameras and laser scanners to obtain corresponding three-dimensional point cloud data. The method for collecting three-dimensional point cloud data can be any known method, which is not limited here.

[0046] S120 , performing registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information.

[0047] In the disclosed embodiment, after collecting the three-dimensional point cloud data, the electronic device can perform registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information.

[0048] Optionally, the preset registration template may be a pre-set registration template. For example, the preset registration template may include a template point cloud similar to a single object point cloud, and topological features of the template point cloud, including but not limited to normals, angles, and other information of adjacent point pairs in the template point cloud, which is not limited here.

[0049] Optionally, the registration process may compare the similarity of features such as normals and angles corresponding to the three-dimensional point cloud data of the object with corresponding features in a preset registration template.

[0050] Optionally, the registration information may be used to indicate the similarity between the object and a preset registration template.

[0051] Specifically, the electronic device may first configure a preset registration template according to the object, and then perform registration processing on the three-dimensional point cloud data according to the preset registration template to obtain corresponding registration information.

[0052] S130. During the registration process, after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud, the successfully registered target three-dimensional point cloud data is removed from the three-dimensional point cloud data, and the registration process is continued until a stop registration condition is reached.

[0053] In an embodiment of the present disclosure, after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud, the electronic device can remove the successfully registered target three-dimensional point cloud data from the three-dimensional point cloud data and continue the registration process until the registration stop condition is reached.

[0054] Optionally, the registration stopping condition may be that the registration process is stopped when the number of successfully registered target three-dimensional point cloud data is equal to the number of the multiple objects.

[0055] Optionally, the registration stopping condition may also be that the registration processing is stopped when the registration processing time reaches a preset processing time. S140, performing posture conversion on the registration information to obtain posture information of the object corresponding to the registration information.

[0056] In the disclosed embodiment, the electronic device may perform posture conversion on the registration information to obtain posture information of the object corresponding to the registration information.

[0057] Optionally, the pose conversion may be estimating the pose information of the object according to the registration information.

[0058] Optionally, the posture information may be information indicating the position and orientation of the object in space.

[0059] Specifically, the electronic device can perform posture conversion on the registration information to obtain posture information of multiple objects. For example, in a workshop scene, the posture information of multiple parts in a material frame in a spatial coordinate system can be obtained.

[0060] Therefore, in the disclosed embodiment, the three-dimensional point cloud data of multiple objects can be obtained, and then the three-dimensional point cloud data is registered according to the preset registration template to obtain the corresponding registration information, and then in the registration process, after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data, the successfully registered target three-dimensional point cloud data is removed from the three-dimensional point cloud data, and the registration process is continued until the registration stop condition is reached, and finally the registration information is converted into a posture to obtain the posture information of the object corresponding to the registration information. Thus, only the three-dimensional point cloud data of the object is collected, and the registration process is performed according to the three-dimensional point cloud data, and the successfully registered target object can be removed, ensuring that all the multiple objects have completed the registration process, thereby avoiding the influence of complex lighting environments on the registration information and improving the accuracy of the posture information of multiple objects.

[0061] Optionally, after S110, the object pose estimation method further includes: preprocessing the three-dimensional point cloud data, wherein the preprocessing includes at least one of downsampling processing and outlier filtering.

[0062] In an embodiment of the present disclosure, the electronic device may pre-process the three-dimensional point cloud data, and the pre-processing includes at least one of downsampling processing and outlier filtering.

[0063] Optionally, preprocessing may be a processing method for improving the quality of three-dimensional point cloud data, wherein the preprocessing includes at least one of downsampling processing and outlier filtering, or any method capable of implementing preprocessing, which is not limited here.

[0064] Specifically, the electronic device can pre-process the three-dimensional point cloud data.

[0065] Figure 2 A schematic diagram of three-dimensional point cloud data provided by an embodiment of the present disclosure is shown.

[0066] like Figure 2 As shown, in a workshop scene, a material frame may include two parts, and the electronic device can collect the three-dimensional point cloud data corresponding to the two parts (the white part in the figure).

[0067] Optionally, the registration process may include a coarse registration process and a fine registration process, wherein the registration process may be an iterative process of adjusting the point cloud data multiple times by using the similarity between the normals, angles and other features of adjacent point pairs of the three-dimensional point cloud data and the corresponding features in the preset registration template so as to reduce the overall deviation between the point cloud and the preset registration template point cloud.

[0068] Optionally, the coarse registration process may be a process of performing preliminary registration on the three-dimensional point cloud data. For example, the coarse registration process may be an existing coarse registration method such as PPF (Point Pair Feature), which is not limited here.

[0069] Optionally, the fine registration process may be a process of further fine registration of the three-dimensional point cloud data. For example, the fine registration process may be an existing fine registration method such as ICP (Iterative Closest Point), which is not limited here.

[0070] Optionally, S120 may specifically include: performing a coarse registration process on the three-dimensional point cloud data according to a preset registration template to obtain preliminary registration information; and performing a fine registration process on the preliminary registration information according to the preset registration template to obtain registration information.

[0071] In the disclosed embodiment, the electronic device may perform a coarse registration process on the three-dimensional point cloud data according to a preset registration template to obtain preliminary registration information.

[0072] Optionally, the preliminary registration information may be information obtained after a coarse registration process.

[0073] Specifically, the electronic device can perform coarse registration processing on the three-dimensional point cloud data according to a preset registration template, such as performing coarse registration processing through a PPF method, and adjust the three-dimensional point cloud data to a degree that is not much different from the template to obtain preliminary registration information.

[0074] Furthermore, the electronic device may perform fine registration processing on the preliminary registration information to obtain registration information.

[0075] Specifically, the electronic device may perform further fine registration processing on the preliminary registration information, for example, perform fine registration processing by an ICP method, to obtain registration information.

[0076] Optionally, the object pose estimation method may further include: if the registration error is less than a first preset threshold, determining that the registration information includes successful registration; if the registration error is greater than or equal to the first preset threshold, determining that the registration information includes unsuccessful registration.

[0077] In some embodiments of the present disclosure, after the electronic device performs the registration process, if the registration error is less than a first preset threshold, it determines that the registration information includes successful registration.

[0078] Optionally, the registration error may be an error value between a registration deviation degree of the three-dimensional point cloud data and a preset deviation threshold.

[0079] Optionally, the registration deviation degree can be used to characterize the deviation degree between the object and the preset registration template after the registration process. For example, the measurement of the registration deviation degree includes but is not limited to the average Euclidean distance between the registered three-dimensional point cloud data and the point cloud data of the preset registration template.

[0080] Optionally, the preset deviation threshold may be a predefined deviation threshold.

[0081] Optionally, the first preset threshold may be a predefined threshold.

[0082] Specifically, after the electronic device performs registration processing, such as iteration, the registration error is determined by the registration deviation degree and a preset deviation threshold. If the registration error is less than a first preset threshold, the registration information is determined to include successful registration.

[0083] In some other embodiments of the present disclosure, after the electronic device performs the registration process, if the registration error is greater than or equal to the first preset threshold, it is determined that the registration information includes unsuccessful registration.

[0084] Specifically, after the electronic device performs registration processing, such as iteration, the registration error is determined by the registration deviation degree and a preset deviation threshold. If the registration error is greater than or equal to the first preset threshold, the registration information is determined to include unsuccessful registration.

[0085] Figure 3 A schematic diagram of successfully registered three-dimensional point cloud data provided by an embodiment of the present disclosure is shown.

[0086] like Figure 3 As shown, the 3D point cloud data (white part) in the upper right corner is successfully registered with the point cloud data (gray part) of the preset registration template.

[0087] Optionally, S130 may specifically include: during the registration process, determining a maximum inclusion area according to the successfully registered three-dimensional point cloud data; and eliminating target three-dimensional point cloud data within the maximum inclusion area.

[0088] In an embodiment of the present disclosure, the electronic device may determine the maximum inclusion area according to the successfully registered three-dimensional point cloud data during the registration process.

[0089] Optionally, the maximum inclusion area may be an area of ​​the three-dimensional point cloud data where the maximum inclusion registration is successful.

[0090] Specifically, during the registration process, the electronic device can determine the maximum inclusion area based on the successfully registered three-dimensional point cloud data. For example, during the registration process, the electronic device can determine the coordinate position data of the successfully registered three-dimensional point cloud data, and thus determine a maximum inclusion area based on the coordinate position data.

[0091] Furthermore, after determining the maximum inclusion area, the electronic device may remove the target three-dimensional point cloud data within the maximum inclusion area.

[0092] Specifically, after determining the maximum inclusion area, the electronic device may eliminate all three-dimensional point cloud data included in the maximum inclusion area, that is, may eliminate the successfully registered target three-dimensional point cloud data.

[0093] Figure 4 A schematic diagram showing a target object removal result provided by an embodiment of the present disclosure is shown.

[0094] like Figure 4 As shown, the electronic device can determine the maximum inclusion area, that is, the cube area in the upper left corner, and eliminate all three-dimensional point cloud data in the maximum inclusion area, that is, eliminate the target object corresponding to the successful registration.

[0095] In this way, the successfully registered target objects can be quickly eliminated.

[0096] Optionally, S130 may further specifically include: during the registration process, determining target three-dimensional point cloud data whose distance between the successfully registered three-dimensional point cloud data and the preset registration template is less than a second preset threshold; and eliminating the target three-dimensional point cloud data.

[0097] In an embodiment of the present disclosure, the electronic device may determine, during the registration process, target three-dimensional point cloud data whose distance between the successfully registered three-dimensional point cloud data and a preset registration template is less than a second preset threshold.

[0098] Optionally, the second preset threshold may be a predefined threshold.

[0099] Optionally, the target three-dimensional point cloud data may be three-dimensional point cloud data whose distance from a preset registration template is less than a second preset threshold.

[0100] Specifically, during the registration process, the electronic device can calculate the three-dimensional point cloud data whose distance between the successfully registered three-dimensional point cloud data and the preset registration template is less than a second preset threshold to obtain the target three-dimensional point cloud data.

[0101] Furthermore, the electronic device eliminates the target object corresponding to the target three-dimensional point cloud data.

[0102] Specifically, the electronic device may remove the target three-dimensional point cloud data after obtaining the target three-dimensional point cloud data.

[0103] In this way, the successfully registered target objects can be accurately eliminated.

[0104] Figure 5 A schematic diagram showing another target object removal result provided by an embodiment of the present disclosure.

[0105] like Figure 5 As shown, the electronic device can determine the target three-dimensional point cloud data, that is, the three-dimensional point cloud data in the upper left corner, and eliminate the target object corresponding to the target three-dimensional point cloud data.

[0106] Optionally, S140 may specifically include: performing posture conversion on the registration information through posture information of a preset registration template, camera parameter information, and parameter information of a robot arm hand-eye calibration to obtain posture information of the object corresponding to the registration information.

[0107] Specifically, after the electronic device performs registration processing on the three-dimensional point cloud data of multiple objects, it can perform posture conversion on the registration information through the posture information of the preset registration template, camera parameter information, and parameter information of the robot arm hand-eye calibration to obtain the posture information of the object corresponding to the registration information.

[0108] Figure 6 A schematic diagram of posture information provided by an embodiment of the present disclosure is shown.

[0109] like Figure 6 As shown, Figure 6 The image includes the pose information of four objects. In the workshop scene, the material frame may include four parts, i.e., four objects. After obtaining the registration information of the objects, the electronic device can perform pose conversion on the registration information through the pose information of the preset registration template, the camera parameter information, and the parameter information of the robot hand-eye calibration to obtain the pose information of the four objects, so as to facilitate the robot arm to grasp according to the pose information.

[0110] Figure 7 A schematic structural diagram of an object posture estimation device provided by an embodiment of the present disclosure is shown.

[0111] In some embodiments of the present disclosure, Figure 7 The object pose estimation device shown can be arranged in an electronic device. Specifically, the electronic device can include but is not limited to mobile terminals such as computer equipment, mobile phones, vehicle-mounted equipment, vehicle controllers, tablet computers, wearable devices, etc.

[0112] like Figure 7 As shown, the object pose estimation device 700 may include a data acquisition module 710 , a registration processing module 720 , an object elimination module 730 and a pose conversion module 740 .

[0113] The data acquisition module 710 can be used to acquire three-dimensional point cloud data of multiple objects.

[0114] The registration processing module 720 can be used to perform registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information.

[0115] The object removal module 730 can be used to remove the successfully registered target three-dimensional point cloud data from the three-dimensional point cloud data after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data during the registration process, and continue the registration process until the registration stop condition is reached.

[0116] The posture conversion module 740 can be used to perform posture conversion on the registration information to obtain the posture information of the object corresponding to the registration information.

[0117] Therefore, in the disclosed embodiment, the three-dimensional point cloud data of multiple objects can be obtained, and then the three-dimensional point cloud data is registered according to the preset registration template to obtain the corresponding registration information, and then in the registration process, after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data, the successfully registered target three-dimensional point cloud data is removed from the three-dimensional point cloud data, and the registration process is continued until the registration stop condition is reached, and finally the registration information is converted into a posture to obtain the posture information of the object corresponding to the registration information. Thus, only the three-dimensional point cloud data of the object is collected, and the registration process is performed according to the three-dimensional point cloud data, and the successfully registered target object can be removed, ensuring that all the multiple objects have completed the registration process, thereby avoiding the influence of complex lighting environments on the registration information and improving the accuracy of the posture information of multiple objects.

[0118] In some embodiments of the present disclosure, the object pose estimation device 700 may further include a data processing module.

[0119] The data processing module can be used to pre-process the three-dimensional point cloud data after collecting the three-dimensional point cloud data of multiple objects, and the pre-processing includes at least one of downsampling processing and outlier filtering.

[0120] In some embodiments of the present disclosure, the registration process may include a coarse registration process and a fine registration process.

[0121] In some embodiments of the present disclosure, the registration processing module 720 may specifically include a first processing unit and a second processing unit.

[0122] The first processing unit can be used to perform coarse registration processing on the three-dimensional point cloud data according to a preset registration template to obtain preliminary registration information.

[0123] The second processing unit can be used to perform fine registration processing on the preliminary registration information according to a preset registration template to obtain registration information.

[0124] In some embodiments of the present disclosure, the object pose estimation device 700 may further include a first determination module and a second determination module.

[0125] The first determination module may be configured to determine that the registration information includes successful registration if the registration error is less than a first preset threshold.

[0126] The second determination module may be configured to determine that the registration information includes unsuccessful registration if the registration error is greater than or equal to a first preset threshold.

[0127] In some embodiments of the present disclosure, the object rejection module 730 may specifically include a first determination unit and a first rejection unit.

[0128] The first determination unit can be used to determine the maximum inclusion area according to the successfully registered three-dimensional point cloud data during the registration process.

[0129] The first elimination unit can be used to eliminate target three-dimensional point cloud data within the maximum inclusion area.

[0130] In some embodiments of the present disclosure, the object rejection module 730 may further specifically include a second determination unit and a second rejection unit.

[0131] The second determination unit may be used to determine, during the registration process, target three-dimensional point cloud data whose distance between the successfully registered three-dimensional point cloud data and a preset registration template is less than a second preset threshold.

[0132] The second elimination unit can be used to eliminate target three-dimensional point cloud data.

[0133] The posture conversion module 740 can be specifically used to perform posture conversion on the registration information through the posture information of the preset registration template, the camera parameter information, and the parameter information of the robot hand-eye calibration to obtain the posture information of the object corresponding to the registration information.

[0134] It should be noted that Figure 7 The object pose estimation device 700 shown can perform Figures 1 to 6 The various steps in the method embodiment shown in the figure are implemented Figures 1 to 6 The various processes and effects in the method embodiment shown are not described in detail here.

[0135] Figure 8 A schematic diagram of the structure of an object pose estimation device provided by an embodiment of the present disclosure is shown.

[0136] In some embodiments of the present disclosure, Figure 8 The object pose estimation device shown may be an intelligent electronic device that a user wants to perform object pose estimation on. The intelligent electronic screen device may include a first device and a second device having an object pose estimation function. Specifically, the electronic device may include but is not limited to mobile terminals such as mobile phones, vehicle-mounted devices, vehicle controllers, tablet computers, wearable devices, smart home devices, etc.

[0137] like Figure 8 As shown, the object pose estimation device may include a processor 801 and a memory 802 storing computer program instructions.

[0138] Specifically, the processor 801 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0139] The memory 802 may include a large capacity memory for information or instructions. By way of example and not limitation, the memory 802 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 802 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 802 may be inside or outside the integrated gateway device. In a particular embodiment, the memory 802 is a non-volatile solid-state memory. In a particular embodiment, the memory 802 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (Electrically Erasable Programmable ROM, EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.

[0140] The processor 801 reads and executes the computer program instructions stored in the memory 802 to perform the steps of the object pose estimation method provided in the embodiment of the present disclosure.

[0141] In one example, the object pose estimation device may further include a transceiver 803 and a bus 804. Figure 8 As shown, the processor 801, the memory 802 and the transceiver 803 are connected via a bus 804 and communicate with each other.

[0142] The bus 804 includes hardware, software, or both. For example, but not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a Memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 804 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0143] The embodiments of the present disclosure also provide a computer-readable storage medium, which may store a computer program. When the computer program is executed by a processor, the processor implements the object pose estimation method provided by the embodiments of the present disclosure.

[0144] The above-mentioned storage medium may, for example, include a memory 802 of computer program instructions, and the above-mentioned instructions may be executed by the processor 801 of the object pose estimation device to complete the object pose estimation method provided by the embodiment of the present disclosure. Optionally, the storage medium may be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium may be a ROM, a random access memory (Random Access Memory, RAM), a compact disc read-only memory (Compact DiscROM, CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0145] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0146] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for estimating an object pose, characterized in that: include: Obtain 3D point cloud data of multiple objects; Performing registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information; In the registration process, after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data, the successfully registered target three-dimensional point cloud data is removed from the three-dimensional point cloud data, and the registration process is continued until a stop registration condition is reached; Performing posture conversion on the registration information to obtain posture information of the object corresponding to the registration information.

2. The method according to claim 1, characterized in that After collecting the three-dimensional point cloud data of the plurality of objects, the method further comprises: The three-dimensional point cloud data is preprocessed, wherein the preprocessing includes at least one of downsampling processing and outlier filtering.

3. The method according to claim 1, characterized in that The registration process includes a coarse registration process and a fine registration process; The registering process of the three-dimensional point cloud data according to the preset registration template to obtain the corresponding registration information includes: Performing rough registration processing on the three-dimensional point cloud data according to the preset registration template to obtain preliminary registration information; The preliminary registration information is finely registered according to the preset registration template to obtain the registration information.

4. The method according to claim 1, characterized in that: The method further comprises: If the registration error is less than a first preset threshold, determining that the registration information includes successful registration; If the registration error is greater than or equal to a first preset threshold, it is determined that the registration information includes unsuccessful registration.

5. The method according to claim 1, characterized in that The step of removing the successfully registered target three-dimensional point cloud data from the three-dimensional point cloud data comprises: During the registration process, determining a maximum inclusion area according to the successfully registered three-dimensional point cloud data; The target three-dimensional point cloud data within the maximum inclusion area is eliminated.

6. The method according to claim 1, characterized in that The step of removing the successfully registered target three-dimensional point cloud data from the three-dimensional point cloud data comprises: In the registration process, determining target three-dimensional point cloud data whose distance between the successfully registered three-dimensional point cloud data and the preset registration template is less than a second preset threshold; The target three-dimensional point cloud data is eliminated.

7. The method according to claim 1, characterized in that The performing posture conversion on the registration information to obtain posture information of the object corresponding to the registration information includes: The registration information is transformed into a posture by using the posture information of the preset registration template, the camera parameter information, and the parameter information of the robot hand-eye calibration to obtain the posture information of the object corresponding to the registration information.

8. An object pose estimation device, characterized in that: include: A data acquisition module, used to obtain three-dimensional point cloud data of multiple objects; A registration processing module, used to perform registration processing on the three-dimensional point cloud data according to a preset registration template to obtain corresponding registration information; An object removal module is used to remove the successfully registered target three-dimensional point cloud data from the three-dimensional point cloud data after obtaining the registration information corresponding to the successfully registered three-dimensional point cloud data during the registration process, and continue the registration process until a stop registration condition is reached; The posture conversion module is used to perform posture conversion on the registration information to obtain the posture information of the object corresponding to the registration information.

9. An object pose estimation device, characterized in that: include: processor; A memory for storing executable instructions; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the object pose estimation method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the object pose estimation method described in any one of claims 1 to 7.

Images