Anchors generation method and device for a three-dimensional scene

Abstract

The application discloses a method and device for generating an anchor point of a three-dimensional scene. The method comprises the following steps: determining a target object in the three-dimensional scene; determining a target observation distance of the target object; determining a target observation visual angle of the target object based on the target observation distance; and the observation position coordinates and observation angle corresponding to the target observation visual angle constitute anchor point information.

Description

Technical Field

[0001] This application relates to the field of 3D modeling technology, and more specifically, to a method and apparatus for generating anchor points in a 3D scene. Background Technology

[0002] Anchor points in a 3D scene are used to mark specific locations or objects within the scene, further providing users with an interactive navigation experience. They allow users to move and explore more easily in a virtual 3D scene using anchor points. Currently, anchor point creation primarily relies on developers manually creating anchor points within development tools. However, this method requires developers to manually create anchor points one by one using development tools, consuming manpower and time. It also limits the number of available anchor points in large 3D scenes, impacting the user experience in large scenes. Summary of the Invention

[0003] In view of the above, this application provides the following technical solution:

[0004] The first aspect of this application provides a method for generating anchor points in a 3D scene, including:

[0005] Identify the target object in the 3D scene;

[0006] Determine the target observation distance of the target object;

[0007] The target observation angle of the target object is determined based on the target observation distance, and the observation position coordinates and observation angle corresponding to the target observation angle constitute the anchor point information.

[0008] In one possible implementation, the method further includes:

[0009] Receive operation information;

[0010] When the operation information corresponds to the anchor point information, the target image is rendered according to the anchor point information and the three-dimensional scene.

[0011] Output the target image.

[0012] In one possible implementation, determining the target object in the 3D scene includes:

[0013] Obtain semantic information from multiple images corresponding to different viewpoints in the three-dimensional scene;

[0014] The target object in the 3D scene is determined based on the semantic information.

[0015] In one possible implementation, obtaining the semantic information of multiple images corresponding to different viewpoints in a 3D scene includes:

[0016] Semantic segmentation is performed on multiple images from different perspectives in a 3D scene to obtain semantic information including the semantically segmented image, semantic category, and coordinates of the semantically segmented object.

[0017] In one possible implementation, determining the target object in the 3D scene based on the semantic information includes:

[0018] The target region of the three-dimensional scene is determined based on the boundary data of the three-dimensional scene;

[0019] At least one object whose coordinates are located in the target area, whose screen occupancy in the image of the three-dimensional scene reaches a set value, and whose semantic category meets preset conditions is identified as the target object.

[0020] In one possible implementation, determining the target observation distance of the target object includes at least one of the following:

[0021] The observation distance corresponding to the proportion of the target object in the three-dimensional scene image within a set range is determined as the target observation distance;

[0022] The target observation distance of the target object is determined based on the size information of the target object and the field of view of the image containing the three-dimensional scene of the target object.

[0023] In one possible implementation, determining the target observation perspective of the target object based on the target observation distance includes:

[0024] Based on the target observation distance, multiple observation images corresponding to the target object are obtained from different perspectives;

[0025] The multiple observation images are subjected to object recognition processing, and the target observation view of the target object is determined based on the recognition processing results.

[0026] In one possible implementation, the step of performing object recognition processing on the multiple observed images and determining the target observation viewpoint of the target object based on the recognition processing results includes:

[0027] The multiple observed images are processed for object recognition to obtain multiple sets of recognition results, each set of recognition results including recognition category and confidence level;

[0028] The viewpoint corresponding to the image with the highest confidence among the observed images whose category is the same as the semantic category of the target object is determined as the target observation viewpoint.

[0029] In one possible implementation, the anchor point information further includes anchor point identification data, which represents the relative positional relationship between the anchor point identification and the target object.

[0030] A second aspect of this application provides an anchor point generation device for a three-dimensional scene, comprising:

[0031] The object determination module is used to determine target objects in a 3D scene;

[0032] The distance determination module is used to determine the target observation distance of the target object;

[0033] An anchor point determination module is used to determine the target observation angle of the target object based on the target observation distance. The observation position coordinates and observation angle corresponding to the target observation angle constitute the anchor point information. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0035] Figure 1 This is a flowchart of a method for generating anchor points in a three-dimensional scene according to an embodiment of this application;

[0036] Figure 2 This is a flowchart illustrating the determination of a target object as disclosed in an embodiment of this application;

[0037] Figure 3 These are images of a target object from different perspectives as disclosed in this embodiment;

[0038] Figure 4 This is a flowchart illustrating the target observation perspective for determining a target object, as disclosed in an embodiment of this application.

[0039] Figure 5 Examples of images of the same object viewed from different perspectives as disclosed in the embodiments of this application;

[0040] Figure 6 This is a schematic diagram illustrating the implementation process of the automatic anchor point generation scheme disclosed in an embodiment of this application;

[0041] Figure 7 This application discloses an anchor point generation device for a three-dimensional scene.

[0042] Figure 8 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. Detailed Implementation

[0043] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0044] The embodiments of this application can be applied to electronic devices. This application does not limit the product form of the electronic device, which may include but is not limited to smartphones, tablets, wearable devices, personal computers (PCs), netbooks, etc., and can be selected according to application requirements.

[0045] Figure 1 This is a flowchart illustrating a method for generating anchor points in a 3D scene according to an embodiment of this application. See also... Figure 1 As shown, methods for generating anchor points in a 3D scene can include:

[0046] Step 101: Identify the target object in the 3D scene.

[0047] The three-dimensional scene can be a scene displayed in different applications and on different devices, such as a scene in a virtual house viewing application, a scene in a game application, or a scene displayed on a smart glasses device. The target object is a virtual object existing in the three-dimensional scene. The target object can be an object identified by matching relevant information input by the user. For example, if the user inputs the voice word "animal" or the text word "animal" in the input box indicating the target object, then the virtual animal in the three-dimensional scene can be identified as the target object. Alternatively, the target object can be an object directly determined by the user's selection operation. For example, if the user directly clicks on an object in the three-dimensional scene with a mouse or finger, then that object can be identified as the target object.

[0048] In implementation, the categories of objects to be focused on in the 3D scene can be determined based on the application purpose and / or user expectations, and these categories can be recorded as required data. For example, in a 3D scene of an art exhibition, the main object category to be focused on are the exhibits; or in a 3D scene of a scenic area, the main object categories to be focused on may include the main entrances of buildings, cultural relics such as tables and chairs; or in a 3D scene of a high-tech park, the main objects to be focused on are monitors, office furniture, and promotional posters.

[0049] There can be one or more target objects in a 3D scene, and there is no fixed limit to their number. For each specific target object, the anchor point generation method disclosed in the embodiments of this application can be used to obtain its anchor point information.

[0050] Step 102: Determine the target observation distance of the target object.

[0051] After identifying the target object, the target observation distance can be determined. This target observation distance can be determined based on pre-set standards, such as the proportion of the target object in the 3D scene.

[0052] In different application scenarios, the criteria used to determine the target observation distance can vary to suit the needs of different 3D scene themes. Of course, due to different user preferences and habits, the criteria used to determine the target observation distance can also be set as a configurable parameter with default values, allowing users to configure it according to their needs.

[0053] Step 103: Determine the target observation angle of the target object based on the target observation distance. The observation position coordinates and observation angle corresponding to the target observation angle constitute the anchor point information.

[0054] After determining the target observation distance, the target observation angle of the target object can be further determined based on the target distance. The target observation angle can be understood as the optimal observation angle of the target object, which should allow the user to clearly see the shape of the target object and make it relatively easy for the user to identify the category of the target object.

[0055] Anchor point information may include the observation position coordinates and observation angle corresponding to the target observation viewpoint, which can be directly determined based on the target observation viewpoint. In other implementations, in addition to the observation position coordinates and observation angle corresponding to the target observation viewpoint, anchor point information may also include other information, such as anchor point display position, anchor point display method, anchor point appearance, etc. This application does not impose fixed limitations on the content included in the anchor point information.

[0056] The anchor point generation method for a 3D scene described in this embodiment can automatically determine the target observation distance and target observation view of the target object after the target object in the 3D scene is determined. Combined with the known coordinate information of the target object, the anchor point information of the target object can be obtained. This anchor point information generation scheme can automatically generate anchor point information in batches. Compared with the traditional scheme of manually creating anchor point information, it is not only more efficient, but also can meet the anchor point generation needs of 3D scenes of different scales.

[0057] In other implementations, the anchor point generation method for a 3D scene may include, in addition to the steps disclosed in the aforementioned embodiments, receiving operation information; rendering a target image based on the anchor point information and the 3D scene when the operation information corresponds to the anchor point information; and outputting the target image.

[0058] The operation information is user-triggered information. For example, if a user directly "clicks" the anchor point of a certain exhibit in the virtual 3D scene of the museum with their finger, or if a user controls the cursor to click the anchor point of the exhibit in the 3D scene of the museum with the mouse, the backend (3D scene server) will render the display image of the exhibit based on the anchor point information and 3D scene data corresponding to the anchor point, and then transmit the rendered image to the terminal device for display, so that the user can view the exhibit from the best viewing angle.

[0059] In the foregoing embodiments, determining the target object in the three-dimensional scene may include: obtaining semantic information of multiple images corresponding to different viewpoints in the three-dimensional scene; and determining the target object in the three-dimensional scene based on the semantic information.

[0060] Images from different perspectives in a 3D scene can be rendered images of the 3D scene, or they can be original photographs used to generate the 3D scene.

[0061] Identifying the target object in a 3D scene, or the core semantic points within that scene, can be understood as the main body to which anchor points are attached. Specifically, semantic segmentation can be performed on multiple images from different viewpoints within the 3D scene to obtain semantic information including the segmented image, semantic category, and coordinates of the segmented object.

[0062] In implementation, images of a 3D scene can be processed based on a semantic segmentation model to obtain images and text information containing detailed semantic information. The image containing semantic information is also known as a semantic segmentation image, which contains multiple semantic segmentation objects. Different categories of semantic segmentation objects are marked with different colors, and the text information records the semantic categories of objects with different color marks. In addition, the text information may also include the coordinates of the semantic segmentation objects.

[0063] Figure 2 This is a flowchart illustrating the determination of a target object as disclosed in an embodiment of this application. (In conjunction with...) Figure 2 As shown, determining the target object in the 3D scene based on the semantic information may include:

[0064] Step 201: Determine the target region of the three-dimensional scene based on the boundary data of the three-dimensional scene.

[0065] Specifically, the target area of ​​the 3D scene can be defined as the central region within a predetermined proportion, such as 70%. The shape of the target area may be the same as or different from the boundary shape of the 3D scene. For example, if the boundary shape of the 3D scene is rectangular, the target area can be rectangular, circular, square, or other shapes. Since the boundary data of the 3D scene is known during construction, the overall area of ​​the 3D scene is also known or can be determined. Therefore, the central region within the 3D scene with a predetermined proportion can be defined as the target area.

[0066] Step 202: Identify at least one object whose coordinates are located in the target area, whose screen occupancy in the image of the three-dimensional scene reaches a set value, and whose semantic category meets the preset conditions as the target object.

[0067] Specifically, ensuring that the image occupies a set percentage of the 3D scene image prevents excessively small objects from being identified as target objects, as these are typically not the focus of user attention. The preset conditions can be the objects the user needs to focus on in the 3D scene, as mentioned earlier. For example, if the target area contains houses, roads, and flowerbeds, and the user identifies buildings as the objects to focus on, then only houses in the target area will be identified as target objects, not roads or flowerbeds. Different preset conditions can also be set for the 3D scene to generate different types of anchor points.

[0068] In other implementations, the process of determining the target object may further include: determining the target main viewpoint of the 3D scene based on the 3D scene image selected by the user. Then, determining one or more objects whose coordinates are located in the target area, whose screen proportion in the 3D scene image reaches a set value, and whose semantic category meets preset conditions as target objects may include: determining one or more objects whose coordinates are located in the target area, whose screen proportion in the image corresponding to the target main viewpoint reaches a set value, and whose semantic category meets preset conditions as target objects.

[0069] Users can select several recommended main view images from the 3D scene images. These recommended main view images include the object of interest (target object) and also represent the optimal viewing angle of the 3D scene. From the optimal viewing angle, the target object in the 3D scene can be clearly displayed. In the implementation, the optimal image can be evaluated from the 3D scene images selected by the user, and the viewpoint corresponding to that image is determined as the target main viewpoint. Figure 3 If the images are from different perspectives of a target object, then based on the user's selection, the perspective corresponding to the image on the right that better displays the animal's form can be determined as the target's main perspective.

[0070] The above content details various implementations for determining the target object, which helps those skilled in the art to better understand and implement the technical solution of this application.

[0071] In the foregoing embodiments, the determination of the target observation distance of the target object can be implemented in different ways. Two of these implementations will be described below.

[0072] In one implementation, the observation distance corresponding to when the proportion of the target object in the image of the 3D scene is within a set range can be determined as the target observation distance. For example, for a target object, the optimal observation distance, i.e., the target observation distance, can be determined by using the proportion of the target object in the image of the scene as approximately 60%.

[0073] In another implementation, considering that the shape and size of the same target object may differ under different field of view angles, this implementation can comprehensively determine the target observation distance based on the size information of the target object and the field of view angle of the 3D scene image containing the target object. The size information characterizes the size of the target object, and the field of view angle matches the presented shape of the target object. The target observation position determined by the combination of these two factors ensures high shape recognition of the target object (making it easier for the user to determine the category of the target object), and that the observed target object is of appropriate size in the image. Specifically, an initial distance can be calculated first based on the size information of the target object, then a field of view angle that satisfies the initial distance and clearly identifies the shape of the target object can be determined based on the field of view angle of the 3D scene image, and finally, the target observation distance including the field of view angle information is determined.

[0074] Understandably, the target observation distance will vary depending on the shape and size of the target object. In applications, the target observation distance can be determined based on pre-set standards, such as the aforementioned proportion of the area in the 3D scene image, or pre-set distance determination logic (related to the aforementioned size information and field of view).

[0075] After determining the target observation distance, the target observation angle of the target object can be determined based on the target observation distance. Figure 4 This is a flowchart illustrating the target observation perspective for determining a target object, as disclosed in an embodiment of this application. (In conjunction with...) Figure 4 As shown, determining the target observation perspective can include:

[0076] Step 401: Based on the target observation distance, obtain multiple observation images corresponding to the target object from different perspectives.

[0077] After determining the target observation distance, you can use the target observation distance as a reference and circle the target object in the virtual 3D scene. During the process, you can randomly capture multiple observation images from different observation angles, such as 10 observation images. These observation images have different perspectives. This can be understood as a person holding a camera and taking multiple photos of an object from different angles. Figure 5 These are example images of the same object viewed from different angles, as disclosed in the embodiments of this application. They can be combined with... Figure 5 Understand the implementation of step 401.

[0078] Step 402: Perform object recognition processing on the multiple observation images, and determine the target observation view of the target object based on the recognition processing results.

[0079] After obtaining multiple observation images, each observation image can be processed separately based on an object recognition model to obtain recognition results. These recognition results may include, but are not limited to, a recognition category and a confidence level. Based on this, object recognition processing is performed on the multiple observation images, and the target observation perspective of the target object is determined based on the recognition processing results. This process may include: performing object recognition processing on the multiple observation images to obtain multiple sets of corresponding recognition results, each set of recognition results including a recognition category and a confidence level; and determining the perspective corresponding to the image with the highest confidence level among the observation images whose recognition category is the same as the semantic category of the target object as the target observation perspective.

[0080] The object recognition model can be any existing or future object recognition model in the field, such as convolutional neural networks, recurrent neural networks, generative adversarial networks, etc., specifically YOLOv8. In one example, multiple observed images are sequentially fed as input to the object recognition model YOLOv8. Using the known semantic category of the target object as a reference, target recognition is performed on these observed images. The YOLO target recognition result includes the object category and a confidence score. The confidence score is a value between 0 and 1, representing the degree of confidence the YOLOv8 model has in the detected target. Based on the confidence score, observed images whose recognition results match the semantic category of the target object are ranked according to their confidence scores. The viewpoint of the image with the highest confidence score is the optimal observation viewpoint for observing the target object, i.e., the target observation viewpoint.

[0081] Once the target observation viewpoint is determined, anchor point information containing the target object's observation position and angle can also be identified. In other implementations, the anchor point information may also include anchor point identifier data, which represents the relative positional relationship between the anchor point identifier and the target object. That is, it specifies where the anchor point identifier should be displayed on the target object, such as above, in front, or to the upper right.

[0082] Figure 6 This is a schematic diagram illustrating the implementation process of the automatic anchor point generation scheme disclosed in an embodiment of this application. (Combined with...) Figure 6 As shown, in a more detailed implementation, the 3D scene image is first input into the semantic segmentation model for semantic segmentation processing, resulting in a semantic segmentation image and text information. These two can be combined with the 3D scene data (including object coordinates) to obtain semantic information. Then, based on the semantic information, the requirement data in the 3D scene data (limiting the categories of objects that users are interested in), the recommended main viewpoint of the 3D image, and the target region obtained based on the 3D scene boundary data, the target object is determined. Based on the determined target object, combined with the 3D scene image, an object recognition model is used to determine the target observation viewpoint of the target object. Once the target observation viewpoint is determined, the anchor point information can be determined.

[0083] This application's solution achieves automatic generation of anchor points in 3D scenes based on semantic segmentation and object recognition models. The process makes full use of existing data in the 3D scene (such as coordinates, boundaries, and images). The automatic generation of anchor points saves manpower. At the same time, since anchor points can be generated in batches, the number is unlimited, which can improve the user experience in large 3D scenes.

[0084] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0085] The methods described in the above-disclosed embodiments of this application are detailed in terms of the methods. The methods of this application can be implemented by various forms of apparatus. Therefore, this application also discloses an apparatus. Specific embodiments are given below for detailed description.

[0086] Figure 7 This application discloses an anchor point generation device for a three-dimensional scene. See also... Figure 7 As shown, the anchor point generation device 70 may include:

[0087] The object determination module 701 is used to determine the target object in the 3D scene.

[0088] The distance determination module 702 is used to determine the target observation distance of the target object.

[0089] Anchor point determination module 703 is used to determine the target observation angle of the target object based on the target observation distance, and the observation position coordinates and observation angle corresponding to the target observation angle constitute anchor point information.

[0090] The anchor point generation device for a 3D scene described in this embodiment can automatically determine the target observation distance and target observation angle of the target object after the target object in the 3D scene is determined. Combined with the known coordinate information of the target object, the anchor point information of the target object can be obtained. This method of generating anchor point information can automatically generate anchor point information in batches. Compared with the traditional method of manually creating anchor point information, it is not only more efficient, but also can meet the anchor point generation needs of 3D scenes of different scales.

[0091] In one implementation, the object determination module may include: a semantic processing module for obtaining semantic information of multiple images corresponding to different viewpoints in the 3D scene; and an object determination submodule for determining the target object in the 3D scene based on the semantic information.

[0092] In one implementation, the semantic processing module can be used to perform semantic segmentation on multiple images from different perspectives in a 3D scene to obtain semantic information including the semantically segmented image, semantic category, and coordinates of the semantically segmented object.

[0093] In one implementation, the object determination submodule may include: a target region determination module, used to determine the target region of the three-dimensional scene based on the boundary data of the three-dimensional scene; and an object selection module, used to determine at least one object whose coordinates are located in the target region, whose screen proportion in the image of the three-dimensional scene reaches a set value, and whose semantic category meets preset conditions as the target object.

[0094] In one implementation, the target region determination module can be used to: determine the middle region of the three-dimensional scene with an area ratio of a set proportion as the target region of the three-dimensional scene, wherein the shape of the target region is the same as or different from the boundary shape of the three-dimensional scene.

[0095] In one implementation, the device may further include: a main viewpoint determination module, used to determine the target main viewpoint of the three-dimensional scene based on the three-dimensional scene image selected by the user; then the object selection module may be specifically used to: determine one or more objects whose coordinates are located in the target area, whose screen proportion in the image corresponding to the target main viewpoint reaches a set value, and whose semantic category meets preset conditions as target objects.

[0096] In one implementation, the distance determination module is specifically used to: determine the observation distance corresponding to when the proportion of the target object in the image of the three-dimensional scene is within a set range as the target observation distance; or, determine the target observation distance of the target object based on the size information of the target object and the field of view of the image of the three-dimensional scene containing the target object.

[0097] In one implementation, the anchor point determination module may include: an observation image acquisition module, used to acquire multiple observation images corresponding to the target object from different perspectives based on the target observation distance; and a perspective determination module, used to perform object recognition processing on the multiple observation images and determine the target observation perspective of the target object based on the recognition processing results.

[0098] In one implementation, the viewpoint determination module can be used to: perform object recognition processing on the multiple observation images to obtain multiple sets of recognition results, each set of recognition results including recognition category and confidence level; and determine the viewpoint corresponding to the image with the highest confidence level among the observation images whose recognition category is the same as the semantic category of the target object as the target observation viewpoint.

[0099] In one implementation, the anchor point information further includes anchor point identification data, which represents the relative positional relationship between the anchor point identification and the target object.

[0100] The anchor point generation device for any of the three-dimensional scenes described in the above embodiments includes a processor and a memory. The object determination module, distance determination module, anchor point determination module, viewpoint determination module, observation image acquisition module, and main viewpoint determination module in the above embodiments are all stored as program modules in the memory. The processor executes the above program modules stored in the memory to realize the corresponding functions.

[0101] The processor contains a kernel, which retrieves the corresponding program modules from memory. One or more kernels can be configured, and the processing of backtracking data can be achieved by adjusting kernel parameters.

[0102] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0103] In an exemplary embodiment, a computer-readable storage medium is also provided, which can be directly loaded into the internal memory of a computer, and contains software code. After being loaded and executed by the computer, the computer program can implement the steps shown in any embodiment of the above-described anchor point generation method for three-dimensional scenes.

[0104] In an exemplary embodiment, a computer program product is also provided, which can be directly loaded into the internal memory of a computer and contains software code. After being loaded and executed by the computer, the computer program can implement the steps shown in any embodiment of the above-described method for generating anchor points in a three-dimensional scene.

[0105] Furthermore, embodiments of this application provide an electronic device. Figure 8This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. See also... Figure 8 As shown, the electronic device 80 includes at least one processor 801, and at least one memory 802 and a bus 803 connected to the processor; wherein the processor and the memory communicate with each other through the bus; the processor is used to call program instructions in the memory to execute the above-mentioned anchor point generation method for the three-dimensional scene.

[0106] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0107] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0108] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0109] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily 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 this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating anchor points in a 3D scene, comprising: Identify the target object in the 3D scene; Determine the target observation distance of the target object; The target observation angle of the target object is determined based on the target observation distance, and the observation position coordinates and observation angle corresponding to the target observation angle constitute the anchor point information.

2. The anchor point generation method for a three-dimensional scene according to claim 1, the method further includes: Receive operation information; When the operation information corresponds to the anchor point information, the target image is rendered according to the anchor point information and the three-dimensional scene. Output the target image.

3. The anchor point generation method for a three-dimensional scene according to claim 1, wherein determining the target object in the three-dimensional scene includes: Obtain semantic information from multiple images corresponding to different viewpoints in the three-dimensional scene; The target object in the 3D scene is determined based on the semantic information.

4. The anchor point generation method for a three-dimensional scene according to claim 3, wherein obtaining semantic information of multiple images corresponding to different viewpoints in the three-dimensional scene includes: Semantic segmentation is performed on multiple images from different perspectives in a 3D scene to obtain semantic information including the semantically segmented image, semantic category, and coordinates of the semantically segmented object.

5. The anchor point generation method for a three-dimensional scene according to claim 4, wherein determining the target object in the three-dimensional scene based on the semantic information includes: The target region of the three-dimensional scene is determined based on the boundary data of the three-dimensional scene; At least one object whose coordinates are located in the target area, whose screen occupancy in the image of the three-dimensional scene reaches a set value, and whose semantic category meets preset conditions is identified as the target object.

6. The anchor point generation method for a three-dimensional scene according to claim 1, wherein determining the target observation distance of the target object includes at least one of the following: The observation distance corresponding to the proportion of the target object in the three-dimensional scene image within a set range is determined as the target observation distance; The target observation distance of the target object is determined based on the size information of the target object and the field of view of the image containing the three-dimensional scene of the target object.

7. The anchor point generation method for a three-dimensional scene according to claim 1, wherein determining the target observation perspective of the target object based on the target observation distance includes: Based on the target observation distance, multiple observation images corresponding to the target object are obtained from different perspectives; The multiple observation images are subjected to object recognition processing, and the target observation view of the target object is determined based on the recognition processing results.

8. The anchor point generation method for a three-dimensional scene according to claim 7, wherein performing object recognition processing on the multiple observation images and determining the target observation viewpoint of the target object based on the recognition processing result includes: The multiple observed images are processed for object recognition to obtain multiple sets of recognition results, each set of recognition results including recognition category and confidence level; The viewpoint corresponding to the image with the highest confidence among the observed images whose category is the same as the semantic category of the target object is determined as the target observation viewpoint.

9. The anchor point generation method for a three-dimensional scene according to claim 1, wherein, The anchor point information also includes anchor point identification data, which represents the relative positional relationship between the anchor point identification and the target object.

10. An anchor point generation device for a three-dimensional scene, comprising: The object determination module is used to determine target objects in a 3D scene; The distance determination module is used to determine the target observation distance of the target object; An anchor point determination module is used to determine the target observation angle of the target object based on the target observation distance. The observation position coordinates and observation angle corresponding to the target observation angle constitute the anchor point information.

Images