Interactive translation system and method based on AR scene recognition and medium
By using scene recognition and runtime container technology in AR devices, the shared translation engine is used to perform interactive translation of virtual scenes, solving the problem of high resource usage during cross-region translation, and improving the portability and user experience of AR devices.
Patent Information
- Application Number
- CN202410919721.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-10
- Publication Date
- 2024-11-12
AI Technical Summary
When existing AR devices translate operation function options and interactive menu items across regions, they consume a lot of resources and cost, which affects the device portability and user experience.
An interactive translation system based on AR scene recognition is adopted to generate AR application content through runtime containers and share a translation engine. Only when needed is activated, the translation engine is required to perform targeted translation, avoiding resource waste and redundancy.
It realizes resource savings for interactive scene recognition and runtime translation in the AR environment, reduces software development and hardware costs, improves user experience, and conforms to the trend of miniaturization and portability of AR devices.
Smart Images

Figure CN118940770A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of AR and translation technology, and in particular to an interactive translation system, method and medium based on AR scene recognition. Background Art
[0002] AR refers to augmented reality. AR can add or remove interactive virtual objects or information generated by computers in real time in the real environment. It is a branch of artificial intelligence. An AR system usually includes technical links such as data processing, 3D registration, display, and human-computer interaction. It obtains real environment information through cameras, combines sensors for positioning tracking and interaction, generates virtual scenes through display devices, and superimposes them on real scenes.
[0003] On the one hand, virtual scenes are usually developed by developers based on their own regional languages, and they usually contain a large number of operation function options and interactive menu items implemented in their own regional languages. When these virtual scenes are transplanted to another region, the difference in regional languages may cause users to be unable to master the operation function options and interactive menu items in the scene. For this reason, manufacturers or developers need to translate these operation function options and interactive menu items to match the native language of users in the current region.
[0004] On the other hand, the large-scale application of AR technology has led to the miniaturization and portability of AR devices, such as head-mounted AR glasses, wearable AR wrist devices, etc. These devices themselves have limited dispatchable resources, and energy consumption costs and battery life must also be considered.
[0005] Therefore, if the complete operation function options and interactive menu items are translated based on the original virtual scene for each VR user in different regions, it will not only be time-consuming and labor-intensive, increasing the development cost of the VR scene, but will also easily lead to the VR device itself requiring a larger storage space, affecting the portability of the VR device and increasing the hardware cost of the VR device. Summary of the invention
[0006] In order to solve the above technical problems, the present invention proposes an interactive translation system, method and medium based on AR scene recognition, which can realize interactive scene recognition and runtime translation in an AR environment with relatively small resource occupation.
[0007] In a first aspect of the present invention, an interactive translation method based on AR scene recognition is proposed. The method is applied to an AR device and comprises the following steps:
[0008] Acquire the focal main field of view of the AR device;
[0009] Determining a next functional operation of a current user of the AR device based on the focal primary field of view;
[0010] Displaying a function menu including a next function operation in the focus main field of view;
[0011] In response to the current user clicking the function menu, translating multiple sub-function items of the function menu, and displaying the translation results in the focus main field of view;
[0012] Each function menu of the AR device includes a plurality of sub-function items to be translated.
[0013] Furthermore, the focus main field of view of the AR device includes a plurality of AR sub-scene areas;
[0014] The determining, based on the focal main field of view, a next functional operation of the current user of the AR device specifically includes:
[0015] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, acquiring an operable function set of the target AR sub-scene area;
[0016] Predicting, based on historical data, the first functional operation of the current user entering the target AR sub-scene area;
[0017] The first functional operation is used as the next functional operation of the current user of the AR device.
[0018] To avoid waste and invalid occupation of memory resources, each AR sub-scene area runs in a runtime container; the runtime container includes a translation engine;
[0019] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container activates the translation engine;
[0020] In response to the current user clicking the function menu, the runtime container translates a plurality of sub-function items of the function menu based on the translation engine;
[0021] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container generates AR application content, where the AR application content includes an AR sub-scene and an AR sub-scene menu function item;
[0022] In response to the current user exiting the target AR sub-scene area, recycling the runtime container corresponding to the target AR sub-scene area.
[0023] To avoid bloated translation engine resources and ineffective redundancy, multiple runtime containers share one translation engine;
[0024] In response to the current user exiting the target AR sub-scene area, the runtime container corresponding to the target AR sub-scene area is recycled, and the connection between the runtime container and the translation engine is disconnected.
[0025] In order to implement the interactive translation method described in the first aspect, in the second aspect of the present invention, an interactive translation system based on AR scene recognition is proposed, the system includes an AR device, the AR device is used to provide an AR application scene, the AR application scene includes an AR sub-scene and functional items to be translated.
[0026] More specifically, the focus main field of view of the AR device includes a plurality of AR sub-scene areas;
[0027] Each of the AR sub-scene areas is generated by the AR device using a runtime container;
[0028] In response to a current user of the AR device entering a first AR sub-scene area in the focus main field of view, the runtime container generates first AR application content, where the first AR application content includes a first AR sub-scene and a first AR sub-scene menu function item;
[0029] In response to an interactive operation of a current user in the first AR sub-scene area, at least one first AR sub-scene menu function item is translated.
[0030] Each of the AR sub-scene areas corresponds to a runtime container;
[0031] Multiple runtime containers share a translation engine;
[0032] The runtime container uses the translation engine to translate at least one first AR sub-scene menu function item.
[0033] In response to the current user exiting the first AR sub-scene area, a first runtime container corresponding to the first AR sub-scene area is recycled, and a connection between the first runtime container and the translation engine is disconnected.
[0034] The above method of the present invention can be automatically implemented based on an electronic device through computer program instructions. The electronic device may include a program storage medium for storing the above computer program instructions.
[0035] Therefore, correspondingly, in the third aspect of the present invention, a computer-readable storage medium is also proposed, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the interactive translation method described in the first aspect are implemented.
[0036] The technical solution of the present invention adopts scene interaction recognition and runtime container generation of AR scene content and triggers targeted calling of translation engine, so as to realize translation of interactive operation function options and interactive menu items only for virtual scenes in AR electronic devices, saving resource calling process; AR application content is generated by runtime container, and in response to the current user exiting the target AR sub-scene area, the runtime container corresponding to the target AR sub-scene area is recovered, and the connection between the runtime container and the translation engine is disconnected, so that interactive scene recognition and runtime translation in AR environment can be realized with less resource occupation, which conforms to the current trend of miniaturization / portability of AR electronic devices, contributes to the large-scale promotion of AR electronic devices and the rapid localization application of virtual scenes, improves user experience and reduces software development cost and hardware cost.
[0037] Further advantages of the present invention will be further reflected in detail in the specific embodiments section in conjunction with the drawings of the specification. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0039] Figure 1 1 is a schematic diagram of the main steps of an interactive translation method based on AR scene recognition according to an embodiment of the present invention;
[0040] Figure 2 Schematic diagram of the main focal field of the AR device involved in the technical solution of the present invention;
[0041] Figure 3 yes Figure 1 A schematic diagram of the interactive translation execution principle of the interactive translation method;
[0042] Figure 4 It is a schematic diagram of the interaction between the runtime container, the translation engine, and the AR sub-scene involved in the technical solution of the present invention;
[0043] Figure 5 It is a schematic diagram of an application scenario of an AR device that implements the technical solution of the present invention;
[0044] Figure 6 It is a partial flowchart of using AR electronic devices to achieve interactive translation. DETAILED DESCRIPTION
[0045] The invention is further described below in conjunction with the accompanying drawings and specific implementation methods.
[0046] AR technology has successfully empowered multiple industries and is not far from people's lives. These industries include machinery manufacturing, automobiles, energy, public safety, medical care, education and other sub-sectors, and AR has begun to occupy a certain scale of use. For example, in the manufacturing industry, in addition to the well-known AR remote collaboration, AR glasses have begun to visualize complex instruction documents, presenting them in 3D in the field of vision, directly replacing the difficult-to-understand paper manuals.
[0047] However, virtual scenes usually contain a large number of operation function options and interactive menu items based on their own regional language (such as English). When these virtual scenes are transplanted to another regional language area (such as Chinese), the difference in regional language may cause users to be unable to master the operation function options and interactive menu items in the scene. For this reason, manufacturers or developers need to translate these operation function options and interactive menu items to match the native language of users in the current region (for example, translating English scenes into Chinese scenes).
[0048] However, if the complete operation function options and interactive menu items are translated based on the original virtual scene for each VR user in different regions, it will not only be time-consuming and labor-intensive, increasing the development cost of the VR scene, but will also easily lead to the VR device itself requiring a larger storage space, affecting the portability of the VR device and increasing the hardware cost of the VR device.
[0049] To this end, the technical solution of the present invention provides multiple specific embodiments to solve one or more technical problems in the above process. It can be understood that in different embodiments of the present invention, each embodiment can solve at least one technical problem, but it is not required that each embodiment solves all technical problems, and each separate embodiment can constitute a separate technical solution.
[0050] See first Figure 1 , Figure 1 It is a schematic diagram of the main steps of an interactive translation method based on AR scene recognition according to an embodiment of the present invention.
[0051] exist Figure 1 In the invention, an interactive translation method based on AR scene recognition is provided, and the method is applied to an AR device.
[0052] AR devices are called augmented reality display devices, which generally refer to head-mounted AR display devices. However, in addition to head-mounted devices, there are also mobile phones, tablets, PCs, car head-up displays (HUD), AR smart interactive devices, etc., which can also be called AR display devices.
[0053] However, the miniaturization and portability of AR devices, such as head-mounted AR glasses and wearable AR wrist devices, are the key to the large-scale implementation of AR applications in the future. Therefore, each embodiment of the present invention is introduced by taking the head-mounted mobile AR device as an example, but it can be understood that the technical solution of the present invention can be applied to other types of AR display devices.
[0054] In AR display devices, the core components include optical and display systems, which can be divided into image source devices and display optical devices. The image source device generates an image and projects the image into the display optical device, which reflects the image into the eye. In addition, the image source device for AR display devices generally refers to a microdisplay, which can also be called a microdisplay plane. The AR display mode is mainly near-eye display: As the name suggests, it is displayed close to the eyes, which is the display mode of AR glasses and AR head-mounted display devices.
[0055] In various embodiments of the present invention, the near-eye display field of view is fully matched with the human eye, and can produce a focal main field of view of 150° to 170° horizontally and 135° to 150° vertically.
[0056] Therefore, see Figure 1 , the steps of the interactive translation method are as follows:
[0057] Acquire the focal main field of view of the AR device;
[0058] Determining a next functional operation of a current user of the AR device based on the focal primary field of view;
[0059] Displaying a function menu including a next function operation in the focus main field of view;
[0060] In response to the current user clicking the function menu, translating multiple sub-function items of the function menu, and displaying the translation results in the focus main field of view;
[0061] Each function menu of the AR device includes a plurality of sub-function items to be translated.
[0062] exist Figure 1 Based on Figure 2-Figure 4 , further introduces various preferred implementation modes of the above embodiments.
[0063] See also Figure 2 , the method first obtains the focus main field of view of the AR device.
[0064] When you start the AR device, the AR device may load the following Figure 2The AR scene includes a 3D book that can be flipped through.
[0065] exist Figure 2 In the example, the field of view where the 3D book is located is the focal main field of view of the AR device.
[0066] The focus main field of view of the AR device includes multiple AR sub-scene areas;
[0067] More specifically, the 3D data can be divided into a left half and a right half. When the user flips through a page of the left half or a page of the right half, he or she enters AR sub-scene 1 or AR sub-scene 2 respectively.
[0068] Based on the superposition of virtual scenes and real scenes using AR technology, taking entering sub-scene 2 as an example, the user flips through a page of a book on the right half, and according to the description of the page, a three-dimensional sub-scene 2 (such as forest-waterfall-white clouds) can be generated in real time; taking entering sub-scene 1 as an example, the user flips through a page of a book on the left half, and according to the description of the page, a three-dimensional sub-scene 1 (such as forest-Monkey King-white clouds) can be generated in real time.
[0069] After entering each sub-scene, the AR scene provides a function menu of corresponding operation buttons. Each function menu contains multiple sub-function items, such as Figure 2 Sub-function item XXX shown.
[0070] In the current scenario, the sub-function item XXX may be displayed as "Pobsanookmonjam", and the local user may not understand its meaning (meaning Pobsanookmonjam), thus affecting the user experience.
[0071] Therefore, see Figure 3 , the method is further implemented as follows:
[0072] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, acquiring an operable function set of the target AR sub-scene area;
[0073] For example, taking entering sub-scene 2 as an example, the user flips through a page of the book on the right half. According to the description of the page, a three-dimensional sub-scene 2 (such as forest-waterfall-white clouds) can be generated in real time. At this time, the target AR sub-scene area is sub-scene 2, and the operable function set includes "Pobsanookmonjam", "mood-elevating" (meaning soaring in the clouds), "WaterfallMoving" (meaning waterfall transfer), etc.
[0074] Assume that "Pobsanookmonjam", "mood-elevating" (meaning flying in the clouds), and "WaterfallMoving" (meaning waterfall transfer) are integrated into a function menu, which is displayed as follows:
[0075] Pobsanookmonjam mood-elevating WaterfallMoving
[0076] Predicting, based on historical data, the first functional operation of the current user entering the target AR sub-scene area;
[0077] The first functional operation is used as the next functional operation of the current user of the AR device.
[0078] As an example, predicting the first functional operation of the current user entering the target AR sub-scene area based on historical data may be:
[0079] The first functional operation of the current user when entering the target AR sub-scene area last time (most recently) is used as the next functional operation of the current user of the AR device;
[0080] Alternatively, the last (last) functional operation of the current user before the last (most recent) exit from the target AR sub-scene area is used as the next functional operation of the current user of the AR device;
[0081] Alternatively, the function operation most frequently used by all users entering the target AR sub-scene area is used as the next function operation of the current user of the AR device.
[0082] Displaying a function menu including a next function operation in the focus main field of view;
[0083] In response to the current user clicking the function menu, multiple sub-function items of the function menu are translated, and the translation results are displayed in the focus main field of view.
[0084] Of course, the user may also click on other function menus (not including the aforementioned next function operation, but including other operations).
[0085] In any case, as long as the current user clicks on the function menu, the multiple sub-function items of the function menu are translated, and the translation results are displayed in the focus main field of view.
[0086] For example, assuming that the next function operation is "Pobsanookmonjam", a function menu including "Pobsanookmonjam" is displayed;
[0087] If the user clicks on the function menu, multiple sub-function items including "Pobsanookmonjam" will be translated, and the translation results will be displayed in the focus main field of view, for example, the function menu is displayed as:
[0088] Boyan Jungle Jump Ride the clouds Waterfall Transfer
[0089] The execution of the above-mentioned “interactive prediction” followed by translation can avoid a large number of pre-translation operations and save device resources.
[0090] In a specific implementation, each of the AR sub-scene areas runs in a runtime container;
[0091] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container generates AR application content, where the AR application content includes an AR sub-scene and an AR sub-scene menu function item;
[0092] The runtime container includes a translation engine;
[0093] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container activates the translation engine;
[0094] In response to the current user clicking the function menu, the runtime container translates a plurality of sub-function items of the function menu based on the translation engine.
[0095] In various embodiments of the present invention, AR application content (including AR scenes, virtual scenes, function menus, etc.) is generated based on the runtime container.
[0096] The runtime container is a pre-developed software stack that runs on the operating system. It does not belong to the operating system. The software stack is provided for AR application development and is required for the development of AR applications. Its role is to provide a runtime environment during the development and operation of AR applications. The runtime environment refers to the environment required for the development of AR applications or the intermediate program code in the AR application development process, such as some software engines and environment variables required for the operation of AR applications or intermediate program codes, and the operation logic required for the operation of AR applications or intermediate program codes. In this embodiment, the software engine is a broad concept, which is a summary of various engines, library functions, etc.
[0097] Each of the AR sub-scene areas runs in a runtime container. Each runtime container is independent of each other, and its life cycle is activated based on the user's interactive operation and destroyed (recycled) based on the user's exit operation.
[0098] Specifically, in response to the current user exiting the target AR sub-scene area, the runtime container corresponding to the target AR sub-scene area is recycled.
[0099] See also Figure 4 To avoid wasting and ineffective use of memory resources, each AR sub-scene area runs in a runtime container; the runtime container includes a translation engine;
[0100] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container activates the translation engine;
[0101] In response to the current user clicking the function menu, the runtime container translates a plurality of sub-function items of the function menu based on the translation engine;
[0102] In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container generates AR application content, where the AR application content includes an AR sub-scene and an AR sub-scene menu function item;
[0103] In response to the current user exiting the target AR sub-scene area, recycling the runtime container corresponding to the target AR sub-scene area.
[0104] To avoid bloated translation engine resources and ineffective redundancy, multiple runtime containers share one translation engine;
[0105] In response to the current user exiting the target AR sub-scene area, the runtime container corresponding to the target AR sub-scene area is recycled, and the connection between the runtime container and the translation engine is disconnected.
[0106] Based on the above embodiments, see Figure 5 .
[0107] Figure 5 It is a schematic diagram of an application scenario of an AR device that implements the technical solution of the present invention.
[0108] based on Figure 5 An AR device is provided, and an interactive translation system based on AR scene recognition is proposed, the system includes Figure 5 The AR device is used to provide an AR application scenario, wherein the AR application scenario includes an AR sub-scenario and functional items to be translated.
[0109] The focus main field of view of the AR device includes multiple AR sub-scene areas;
[0110] Each of the AR sub-scene areas is generated by the AR device using a runtime container;
[0111] In response to a current user of the AR device entering a first AR sub-scene area in the focus main field of view, the runtime container generates first AR application content, where the first AR application content includes a first AR sub-scene and a first AR sub-scene menu function item;
[0112] In response to an interactive operation of a current user in the first AR sub-scene area, at least one first AR sub-scene menu function item is translated.
[0113] Each of the AR sub-scene areas corresponds to a runtime container;
[0114] Multiple runtime containers share a translation engine;
[0115] The runtime container uses the translation engine to translate at least one first AR sub-scene menu function item.
[0116] Figure 6 It is a partial flowchart of using AR electronic devices to achieve interactive translation.
[0117] After the AR device is started, the current user enters the first AR sub-scene area in the focus main field of view, and the first runtime container generates first AR application content, where the first AR application content includes the first AR sub-scene and the first AR sub-scene menu function item;
[0118] At this time, if the current user performs an interactive operation in the first AR sub-scene area, at least one first AR sub-scene menu function item is translated.
[0119] As a specific example, the current user performs an interactive operation in the first AR sub-scene area, including:
[0120] Determining a next functional operation of a current user of the AR device based on the focal primary field of view;
[0121] Displaying a function menu including a next function operation in the focus main field of view;
[0122] The current user clicks on the function menu;
[0123] Of course, of course, the user may also click on other function menus (not including the aforementioned next function operation, but including other operations).
[0124] In any case, as long as the current user clicks on the function menu, the multiple sub-function items of the function menu are translated, and the translation results are displayed in the focus main field of view.
[0125] Next, if the current user exits the first AR sub-scene area, the first runtime container corresponding to the first AR sub-scene area is recovered, and the connection between the first runtime container and the translation engine is disconnected.
[0126] By adopting scene interaction recognition and runtime container to generate AR scene content and trigger targeted calling of translation engine, the technical solution of the present invention realizes the translation of interactive operation function options and interactive menu items only for virtual scenes in AR electronic devices, saving the resource calling process; AR application content is generated by runtime container, and in response to the current user exiting the target AR sub-scene area, the runtime container corresponding to the target AR sub-scene area is recovered, and the connection between the runtime container and the translation engine is disconnected, so that interactive scene recognition and runtime translation in AR environment can be realized with less resource occupation, which conforms to the current trend of miniaturization / portability of AR electronic devices, contributes to the large-scale promotion of AR electronic devices and the rapid localization application of virtual scenes, improves user experience while reducing software development costs and hardware costs.
[0127] It should be noted that in the specific implementation of this application, if user-related data is involved, when the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.
[0128] The foregoing has shown and described the method embodiments and system of the present invention, but it is understandable to those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.
[0129] The above method of the present invention can be automatically executed by an electronic device through a computer program, and the computer program is stored in a computer-readable storage medium. The computer program can be executed by at least one processor so that the at least one processor performs the steps of an interactive translation method based on AR scene recognition described in the first aspect.
[0130] In addition, the embodiments of the present invention are divided into embodiments in the form of method flow, embodiments in the form of system functional architecture, and embodiments in the form of computer program code pseudo language. It is understood that, unless otherwise specified, the same or corresponding steps between the embodiments in different forms correspond to each other. For example, the "XX unit, used to implement XX function" in the embodiment in the form of system functional architecture actually corresponds to the "XX step, used to implement XX function" in the embodiment in the form of method flow.
[0131] Therefore, after the embodiment in the form of method flow is elaborated in detail, the corresponding embodiment in the form of system function architecture does not need to be repeated again, but those skilled in the art can directly and unambiguously determine that the embodiment in the form of system function architecture also corresponds to all the contents of the embodiment in the form of method flow. The same is true for the embodiment in the form of computer program code pseudo language.
Claims
1. An interactive translation method based on AR scene recognition, the method being applied to an AR device, characterized in that: The method comprises: Acquire the focal main field of view of the AR device; Determining a next functional operation of a current user of the AR device based on the focal primary field of view; Displaying a function menu including a next function operation in the focus main field of view; In response to the current user clicking the function menu, translating multiple sub-function items of the function menu, and displaying the translation results in the focus main field of view; Each function menu of the AR device includes a plurality of sub-function items to be translated.
2. The interactive translation method based on AR scene recognition as claimed in claim 1, characterized in that: The focus main field of view of the AR device includes multiple AR sub-scene areas; The determining, based on the focal main field of view, a next functional operation of the current user of the AR device specifically includes: In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, acquiring an operable function set of the target AR sub-scene area; Predicting, based on historical data, the first functional operation of the current user entering the target AR sub-scene area; The first functional operation is used as the next functional operation of the current user of the AR device.
3. The interactive translation method based on AR scene recognition as claimed in claim 2, characterized in that: Each of the AR sub-scene areas runs in a runtime container; In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container generates AR application content, where the AR application content includes an AR sub-scene and an AR sub-scene menu function item.
4. The interactive translation method based on AR scene recognition as claimed in claim 3, characterized in that: In response to the current user exiting the target AR sub-scene area, recycling the runtime container corresponding to the target AR sub-scene area.
5. The interactive translation method based on AR scene recognition as claimed in claim 3, characterized in that: The runtime container includes a translation engine; In response to the current user entering one of the target AR sub-scene areas in the focus main field of view, the runtime container activates the translation engine; In response to the action of the current user clicking on the function menu, the runtime container translates a plurality of sub-function items of the function menu based on the translation engine.
6. The interactive translation method based on AR scene recognition as claimed in claim 3, characterized in that: A plurality of the runtime containers share a translation engine; In response to the current user exiting the target AR sub-scene area, the runtime container corresponding to the target AR sub-scene area is recycled, and the connection between the runtime container and the translation engine is disconnected.
7. An interactive translation system based on AR scene recognition, the system comprising an AR device, the AR device being used to provide an AR application scene, the AR application scene comprising an AR sub-scene and a function item to be translated, characterized in that: The focus main field of view of the AR device includes multiple AR sub-scene areas; Each of the AR sub-scene areas is generated by the AR device using a runtime container; In response to a current user of the AR device entering a first AR sub-scene area in the focus main field of view, the runtime container generates first AR application content, where the first AR application content includes a first AR sub-scene and a first AR sub-scene menu function item; In response to an interactive operation of a current user in the first AR sub-scene area, at least one first AR sub-scene menu function item is translated.
8. The interactive translation system based on AR scene recognition as claimed in claim 7, characterized in that: Each of the AR sub-scene areas corresponds to a runtime container; Multiple runtime containers share a translation engine; The runtime container uses the translation engine to translate at least one first AR sub-scene menu function item.
9. The interactive translation system based on AR scene recognition as claimed in claim 8, characterized in that: In response to the current user exiting the first AR sub-scene area, a first runtime container corresponding to the first AR sub-scene area is recycled, and a connection between the first runtime container and the translation engine is disconnected.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the interactive translation method according to any one of claims 1 to 6 are implemented.
