Projector and interactive control method

Abstract

A projector and its interactive control method. The interactive control method of the projector includes: acquiring unique identification information and interactive action information of a target object; inputting the unique identification information and interactive action information of the target object into a preset interactive model to obtain the interactive result corresponding to the interactive action information; and controlling the projection content of the projector according to the interactive result.

Description

Technical Field

[0001] This invention relates to the field of projection technology, and more particularly to a projector and its interactive control method. Prior Technology

[0002] Projectors can be widely used in homes, offices, schools, and entertainment venues. They can project images or videos onto display devices such as screens or walls. They can also be connected to portable or non-portable terminals through different interfaces to project the content displayed on the terminal onto the display device, thereby meeting the user's display needs.

[0003] When a projector projects content, users can create, annotate, or modify the projected content according to their needs. In existing technologies, users can use a terminal connected to the projector, or devices such as laser pointers or remote controls that wirelessly communicate with the projector or the projected content, to create, modify, or annotate the projected content.

[0004] However, when creating, modifying, or annotating projected content using existing technologies, users need to be physically present at the projector or have access to a device such as a laser pointer or remote control that wirelessly communicates with the projector or the display device showing the projected content. This makes the process complex and unsuitable for collaborative creation, modification, or annotation by multiple users. Therefore, improving the convenience and accuracy of creating, modifying, or annotating projected content has become a pressing technical challenge. Summary of the Invention

[0005] This invention provides a projector and its interactive control method to solve the technical problem in the prior art that it is impossible to conveniently and accurately create, modify or mark the projected content of the projector.

[0006] According to one aspect of the present invention, a projector interactive control method is provided, comprising: acquiring image information and / or voice information of each actual participating object in real time; determining an action identifier of each actual participating object based on the image information and / or voice information of each actual participating object; identifying the actual participating object whose action identifier is a preset identifier as the target object; acquiring unique identifier information and interactive action information of the target object; inputting the unique identifier information and interactive action information of the target object into a preset interaction model to obtain an interaction result corresponding to the interactive action information; controlling the projection content of the projector based on the interaction result. The projector interactive control method further comprises: acquiring image information and / or voice information of the target object in real time; adjusting the projection content based on the image information and / or voice information of the target object.

[0007] In some embodiments, the projector interactive control method further includes: acquiring unique identifier information and interactive feature information of each initial participating object; constructing a model training database that corresponds one-to-one between the unique identifier information and the interactive feature information based on the unique identifier information and the interactive feature information of each initial participating object; and training a deep learning model through the model training database to obtain the preset interactive model.

[0008] In some embodiments, the projector interactive control method further includes: updating the model training database based on the unique identifier information of the target object and the interactive action information.

[0009] In some implementations, updating the model training database based on the unique identifier information of the target object and the interaction action information includes: determining whether the unique identifier information of the target object exists in the model training database; if so, updating the interaction feature information corresponding to the unique identifier information of the target object in the model training database based on the unique identifier information of the target object and the interaction action information.

[0010] In some implementations, updating the model training database based on the unique identifier information and the interaction action information of the target object further includes: if the unique identifier information of the target object does not exist in the model training database, then using the interaction action information as interaction feature information corresponding to the unique identifier information of the target object; and adding the unique identifier information and the interaction feature information of the target object to the model training database.

[0011] In some implementations, obtaining the unique identification information and interactive action information of the target object includes: instantly acquiring image information and / or voice information of the target object; and determining the unique identification information and interactive action information of the target object based on the image information and / or voice information of the target object.

[0012] In some implementations, determining the interactive action information of the target object based on the image information and / or voice information of the target object includes: determining the actual action information and action reference information of the target object based on the image information and / or voice information of the target object; and determining the interactive action information of the target object based on the actual action information and the action reference information.

[0013] According to another aspect of the present invention, a projector is provided, comprising: a data processing module, an image acquisition module, a voice input module, and a projection output module; the image acquisition module is used to acquire image information; the voice input module is used to acquire voice information; the projection output module is used to output projection content; the data processing module is connected to the image acquisition module, the voice input module, and the projection output module respectively; the data processing module is used to acquire the image information and the voice information, execute the above-described projector interactive control method, and control the projection output module to output projection content.

[0014] In some embodiments, the image acquisition module includes multiple cameras arranged in an array; each camera acquires image information from different angles around the projector.

[0015] In some embodiments, the projector further includes a wireless communication module; the data processing module interacts with a remote terminal and / or server via the wireless communication module.

[0016] The technical solution of this invention obtains the unique identifier information and interaction action information of a target object, and inputs this information into a preset interaction model to obtain an interaction result corresponding to the interaction action information of the target object. This interaction result is determined based on the personalized characteristics of the target object and the actual interaction action, ensuring high accuracy. This results in highly accurate projection content when controlling the projector's projection content based on the interaction result. Thus, without the need for a terminal device, the projector can directly create, mark, or modify the projected content using the unique identifier information and interaction action information of the target object, thereby improving the projector's interactive convenience. Furthermore, considering the personalized interaction characteristics of the target object, the accuracy of the determined projection content is ensured, ultimately enhancing the user experience.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Simple Explanation of the Diagram

[0018] To make the above and other objects, features, advantages and embodiments of this disclosure more apparent and understandable, the accompanying drawings are described below: Figure 1 is a structural block diagram of a projector provided in an embodiment of the present invention; Figure 2 is a schematic diagram of the structure of a projector provided in an embodiment of the present invention; Figure 3 is a structural block diagram of another projector provided in an embodiment of the present invention; Figure 4 is a flowchart of a projector interactive control method provided in Embodiment 2 of the present invention; Figure 5 is a schematic diagram of the projected content displayed by a projector on a display device according to Embodiment 2 of the present invention; Figure 6 is a schematic diagram of the projected content displayed by a projector on another display device provided in Embodiment 2 of the present invention; Figure 7 is a flowchart of a projector interactive control method provided in Embodiment 3 of the present invention; Figure 8 is a flowchart of a projector interactive control method provided in Embodiment 4 of the present invention; Figure 9 is a flowchart of a projector interactive control method provided in Embodiment 5 of the present invention; Figure 10 is a flowchart of a projector interactive control method provided in Embodiment 6 of the present invention. Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such usage can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] The acquisition, storage, use, and processing of data in the technical solution of this invention all comply with the relevant provisions of relevant laws and regulations.

[0022] Example 1: Figure 1 is a structural block diagram of a projector provided in Example 1 of the present invention. As shown in Figure 1, the projector 100 includes a data processing module 10, an image acquisition module 20, a voice input module 30, and a projection output module 40. The image acquisition module 20 is used to acquire image information; the voice input module 30 is used to acquire voice information; the projection output module 40 is used to output projection content. The data processing module 10 is connected to the image acquisition module 20, the voice input module 30, and the projection output module 40 respectively. The data processing module 10 is used to acquire image information and / or voice information, execute the projector interactive control method provided in this embodiment of the present invention, and control the projection output module 40 to output projection content.

[0023] The image acquisition module 20 may include a camera or similar device to acquire image information around the projector 100 and send the acquired image information to the data processing module 10 for processing. The voice input module 30 may include a microphone or similar device to pick up voice information around the projector 100 and send the picked-up voice information to the data processing module 10. The data processing module 10 can process the received image information and / or voice information to determine the unique identifier information and interaction action information of the target object, input the unique identifier information and interaction action information of the target object into a preset interaction model, obtain the interaction result corresponding to the interaction action information of the target object, and then control the projection output module 40 to output the projected content to the display device according to the interaction result.

[0024] Understandably, a preset interaction model can be obtained by training a deep learning model. By inputting the unique identifier information and interaction feature information of the target object into the deep learning model for training, the deep learning model can learn the personalized feature information of the target object, thereby obtaining the preset interaction model. When the target object generates interactive action information, the preset interaction model can be used to determine the interaction result corresponding to the interactive action information, thus enabling more accurate identification of the interactive action information of the target object.

[0025] In an optional embodiment, the image information acquired by the image acquisition module 20 may include facial feature information of the target object associated with the unique identification information of the target object, and the voice information picked up by the voice input module 30 may include voiceprint feature information associated with the unique identification information of the target object, so that after the data processing module 10 processes the image information and / or voice information, the facial feature information and voiceprint feature information of the target object can be determined, thereby determining the unique identification information of the target object.

[0026] It should be noted that in this embodiment of the invention, the data processing module 10 is connected to the image acquisition module 20, the voice input module 30, and the projection output module 40, respectively. That is, the data processing module 10 can be electrically connected to the image acquisition module 20, the voice input module 30, and the projection output module 40 via wires, or it can be communicatively connected to the image acquisition module 20, the voice input module 30, and the projection output module 40 via a wireless communication protocol. Furthermore, the connection methods of the data processing module 10 with the image acquisition module 20, the voice input module 30, and the projection output module 40 can be the same or different, and can be designed according to actual needs. This embodiment of the invention does not impose specific limitations on this.

[0027] This embodiment employs an image acquisition module to collect image information and a voice input module to acquire voice information. This allows the data processing module to determine the unique identifier and interactive action information of the target object based on the image and / or voice information. This unique identifier and interactive action information are then input into a preset interaction model to determine the interaction result corresponding to the interactive action information of the target object. Based on this interaction result, the projection output module can accurately output the projected content. Thus, accurate interaction between the target object and the projector can be achieved without the need for other devices, simplifying the projector's interactive operation, improving the projector's interactive convenience, and ultimately enhancing the user experience.

[0028] As shown in Figure 2, the projector 100 may also include a housing 101. The cameras 21 of the image acquisition module 20 may be arranged in a ring along the edge of the housing 101, so that each camera can acquire image information from different angles. This allows the image acquisition module 20 to acquire image information from a 360° angle around the projector 100. In this way, the image acquisition module 20 can identify the image information of target objects located in any position of the projector 100, so that target objects located in any position of the projector 100 can accurately interact with the projector.

[0029] Optionally, referring to Figure 1, the voice input module 30 may include multiple microphones arranged in a ring to pick up voice information from various directions, so that the voice of the target object can be picked up from any position, thereby improving the accuracy of voice recognition and the convenience of voice interaction.

[0030] Optionally, Figure 3 is a structural block diagram of another projector provided in Embodiment 1 of the present invention. As shown in Figure 3, the projector 100 may further include a wireless communication module 50. In this case, the data processing module 10 can interact with the remote terminal 51 and / or server 52 through the wireless communication module 50.

[0031] The wireless communication module 50 may include a Bluetooth module, a WIFI module, etc. As long as the projector 100 can achieve wireless communication with the remote terminal 51 and / or server 52, the embodiments of the present invention do not make specific limitations in this regard.

[0032] Specifically, the data processing module 10 can interact with the remote terminal 51 via the wireless communication module 50. That is, the data processing module 10 can send at least one of the image information collected by the image acquisition module 20, the voice information acquired by the voice input module 30, and the projection content output by the control projection output module 40 to the remote terminal 51 via the wireless communication module 50, so that the participating objects can understand the on-site situation of the projector and watch the projected content through the remote terminal. At the same time, the data processing module 10 can also receive the image information and / or voice information collected by the remote terminal 51, as well as the projection content sent by the remote terminal 51, so that the participating objects at the remote terminal can interact with the projector 100.

[0033] The data processing module 10 can also interact with the server 52 via the wireless communication module 50. This server 52 can be a cloud server, solving the problems of high management difficulty and weak business scalability between physical hosts and virtual private servers (VPS). For example, the data processing module 10 can upload received image and / or voice information, as well as its output projection content, to the server 52 for storage, facilitating subsequent retrieval; alternatively, the data processing module 10 can download corresponding projection content and historically uploaded data from the server 52, thereby meeting the projection needs of the projector while making the projector's projection methods more diverse.

[0034] In other alternative embodiments, the remote terminal 51 can also interact with the server 52 to enable the remote terminal 51 to upload its terminal data to the server 52 and download corresponding data from the server 52. The specific communication method can be designed according to actual needs, and the embodiments of the present invention do not impose specific limitations on this.

[0035] Based on the above embodiments, optionally, referring to FIG3, the projector 100 may further include a storage module 60 connected to the data processing module 10, for storing image information acquired by the image acquisition module 20, voice information acquired by the voice input module 30, and unique identification information and interactive action information generated based on the image information and / or voice information, so that when the data processing module 10 performs signal processing, the content stored in the storage module 60 can be directly called, simplifying the signal processing process and improving the signal processing efficiency.

[0036] Example 2: Figure 4 is a flowchart of a projector interactive control method provided in Example 2 of the present invention. This example is applicable to realizing personalized interaction with the projected content of a projector without the aid of external devices. This method can be executed by a projector interactive control device, which can be implemented in hardware and / or software form, and can be configured in the data processing module of the projector. As shown in Figure 4, the method includes: step S110, obtaining the unique identification information of the target object and the interaction action information.

[0037] It is understandable that the unique identification information of the target object can be unique and deterministic identification information of the target object, that is, different target objects have different unique identification information, and the unique identification information can exist in an encoded way; the interactive action information of the target object can be the interactive information such as gestures and voice emitted by the target object when interacting with the projector. For example, the interactive action information can be the gesture action information of the target object. By acquiring the gesture action information of the target object in real time, it is possible to determine the image drawn by the target object, the text written by the target object, etc. Alternatively, the interactive action information can also be the voice information of the target object. By acquiring the sound emitted by the target object, it is possible to know the text, image, etc. expressed by the target object.

[0038] The projector may include an image acquisition module and a voice input module. The image acquisition module can acquire image information of the target object, and the voice input module can acquire voice information of the target object. Based on the image information and / or voice information, the unique identifier and interactive action information of the target object can be determined. In other optional embodiments, the projector may also communicate with a remote terminal. Image information can be acquired using a camera or similar device in the remote terminal, and voice information can be acquired using a microphone or similar device in the remote terminal. The specific implementation can be designed according to actual needs, and this embodiment of the invention does not impose specific limitations on this.

[0039] In an exemplary embodiment, after obtaining image information of the target object, facial feature information of the target object can be determined based on the image information. This facial feature information is a unique and definitive identifier of the target object. Therefore, the unique identifier of the target object can be determined based on the facial feature information. Alternatively, after obtaining voice information of the target object, voiceprint feature information of the target object can be determined based on the voice information. This voiceprint feature information can also serve as a unique and definitive identifier of the target object. Therefore, the unique identifier of the target object can also be determined based on the voiceprint feature information. Alternatively, after determining the facial feature information of the target object based on the image information and the voiceprint feature information of the target object based on the voice information, the unique identifier of the target object can be determined by combining the facial feature information and the voiceprint feature information.

[0040] Correspondingly, by acquiring image information of the target object in real time through the image acquisition module, the gesture coordinates of the target object at each moment can be determined based on the image information acquired at each moment. Based on the gesture coordinates at each moment, the gesture action of the target object can be determined, and thus the interactive action information of the target object can be determined based on the gesture action. For example, if the finger or hand holding a laser pointer moves along direction A, the interactive action information of the target object can be determined as drawing a straight line along direction A based on the image information of the target object. Alternatively, by acquiring voice information of the target object in real time through the voice input module, the corresponding text information can be determined, and this text information can be used as the interactive action information of the target object. For example, if the voice input module acquires the voice of the target object, "Insert content A in the center area", then the interactive action information of the target object is "Insert content A in the center area". Alternatively, the interactive action information of the target object can be determined by combining the image information acquired by the image acquisition module and the voice information acquired by the voice input module. In this case, the target object can simultaneously perform gesture actions and voice output. For example, the target object's finger or handheld laser pointer points to a certain position on the display surface of the display device used to display the projected content of the projector, and at the same time, the target object outputs the voice message "Insert content A here". In this case, inserting content A at the position pointed to by the target object is the interactive action information of the target object.

[0041] It is understandable that when target objects have different sizes and different distances from the display device (such as a display screen, background wall, or screen) displaying the projected content, the interactive action information presented by the same gesture amplitude of the target object will differ, thus hindering the accurate identification of the target object's interactive action information. In this case, a reference object can be determined according to actual needs. Based on the action and / or size of the reference object, the relative action amplitude of the target object's gesture amplitude with respect to the action and / or size of the reference object can be determined, and this relative action amplitude can be used as the interactive action information of the target object.

[0042] In an optional embodiment, determining the interactive action information of the target object based on the image information and / or voice information of the target object may include: determining the actual action information and action reference information of the target object based on the image information and / or voice information of the target object; and determining the interactive action information of the target object based on the actual action information and action reference information.

[0043] As a feasible embodiment, when determining the motion reference information and actual motion information of the target object based on the image information of the target object, the motion reference information can be the reference object size information of a static reference object in the acquired image information, or the motion reference information can be the reference object size information and / or reference object motion information of a dynamic reference object in the acquired image information. The actual motion information can be the actual hand gesture of the target object, and the amplitude of the hand gesture of the target object can be determined based on the actual hand gesture.

[0044] When the motion reference information is the size information of a dynamic reference object, this dynamic reference object can be a reference object that moves in response to the gesture of the target object, such as the palm of the target object. In this case, based on the proportional relationship between the amplitude of the gesture and the size of the palm, the position and size information of the gesture on the display surface of the display device can be determined accordingly. This position and size information can then be used as interactive motion information. Thus, by using motion reference signals and actual motion information, the interactive motion information of the target object can be determined more accurately, reducing the impact of differences in the size of the target object and the distance between the target object and the display device on the interactive motion information.

[0045] It is understood that the above description is only exemplified by the fact that both motion reference information and actual motion information are determined by image information or voice information. In other feasible embodiments of the present invention, motion reference information can be determined based on image information, and actual motion information can be determined based on voice information; or, motion reference information can be determined based on voice information, and actual motion information can be determined based on image information.

[0046] Step S120: Input the unique identifier information and interaction action information of the target object into the preset interaction model, and obtain the interaction result corresponding to the interaction action information.

[0047] The preset interaction model can be understood as a method for identifying unique identifier information and determining personalized feature information corresponding to the unique identifier based on the identified unique identifier information. Based on the personalized feature information, a personalized interaction result corresponding to the interaction action information is determined. This allows for the targeted determination of the personalized features of the target object and the accurate determination of the interaction result corresponding to the interaction action information, thereby improving the accuracy of the determined interaction result.

[0048] Specifically, by inputting the unique identifier and interaction action information of the target object into a preset interaction model, the information processing module of the preset interaction model first determines the interaction feature information corresponding to the unique identifier. This interaction feature information may include information related to the personalized features corresponding to the unique identifier. Then, the feature processing module of the preset interaction model compares the interaction action information with the interaction feature information, determining the interaction feature information whose error with the interaction action information is within a preset tolerance range. Finally, the result output module of the preset interaction model determines the interaction result corresponding to the determined interaction feature information, i.e., the interaction result corresponding to the interaction action information. Thus, by inputting the unique identifier and interaction action information of the target object into the preset interaction model, personalized interaction results for the target object can be determined specifically, improving the accuracy of the obtained interaction results.

[0049] In an exemplary embodiment, when the interaction action information of a target object is determined based on its image information, the image information may include the target object's gesture action. Since the movement direction and amplitude of gesture actions differ among different target objects, it is necessary to determine the interaction result corresponding to the target object's interaction action information based on the personalized characteristics of each target object. For example, when a target object needs to draw a straight line of length d on the projected screen, the actual gesture is to move a distance d' from left to right. Based on image information, the direction and distance of the target object's actual gesture are identified as the target object's interactive action information. After inputting the target object's unique identifier and interactive action information into a preset interaction model, personalized feature information corresponding to the target object's unique identifier can be determined. Then, the interactive action information determined based on the target object's actual gesture can be compared with the target object's personalized feature information. This allows us to determine that the content the target object's interactive action information expresses is drawing a straight line of length d on the projected screen, thus enabling the targeted determination of a personalized interaction result corresponding to the target object's interactive action information.

[0050] In another exemplary embodiment, when the interactive action information of the target object is determined based on the voice information of the target object, the voice information may include the pitch, intensity, duration, timbre, and language expression of the target object. Since the pitch, intensity, duration, and timbre of different target objects are different, and the language expression of target objects in different regions are different, for example, the voices sent by target objects in different regions may carry different regional dialects, after inputting the unique identification information and interactive action information of the target object into the preset interaction model, the regional dialect information corresponding to the unique identification information of the target object can be determined. Based on the regional dialect information of the target object, the text information to be expressed by each sound in the interactive action information of the target object can be determined. Based on the text information, the personalized interaction result corresponding to the interactive action information of the target object can be determined.

[0051] Step S130: Control the projection content of the projector based on the interaction results.

[0052] The projector can project content such as images, text, and videos onto the display device. The interaction result of the identified target object is the content that the target object wants to express. This content is then converted into text, images, videos, etc., and displayed on the display device as the projector's projection content.

[0053] In an exemplary embodiment, as shown in FIG5, when image information of a target object is acquired based on an image acquisition module, and the gesture action of the target object is determined to be a movement of a distance d' from left to right based on the acquired image information, the position coordinates of each key point in the gesture action of the target object in the projector coordinate system and the distance between each key point can be determined based on the image information. For example, the position coordinates of the starting position point in the projector coordinate system (x1', y1') (not shown), the position coordinates of the ending position point in the projector coordinate system (x2', y2') (not shown), and the actual distance d' between the starting position point and the ending position point (not shown). The position coordinates of each key point in the gesture action in the projector coordinate system and the actual distance between each key point are used as the interaction action information of the target object. After the identity information and interaction action information are input into the preset interaction model, the personalized interaction result corresponding to the interaction action information of the target object can be determined. That is, the position coordinates (x1´, y1´) of the starting position point in the projector coordinate system are converted to the position coordinates (x1, y1) of the starting position point in the display device coordinate system, the position coordinates (x2´, y2´) of the ending position point in the projector coordinate system are converted to the position coordinates (x2, y2) of the ending position point in the display device coordinate system, and the actual distance d´ between the starting position point and the ending position point is converted to the length d displayed on the display device. Based on the interaction result, the projection content of the projector can be determined to be a line segment of length d displayed on the display device, and the starting position point of the line segment is (x1, y1) and the ending position point is (x2, y2).

[0054] In another exemplary embodiment, as shown in Figure 6, when the voice information of the target object is obtained based on the voice input module, and the voice of the target object is determined to be "insert content A in the central area" based on the obtained voice information, then each character in the voice content of the target object can be determined based on the voice information, and the determined characters can be used to determine the interactive action information of the target object. After inputting the unique identification information and interactive action information of the target object into the preset interaction model, a personalized interaction result corresponding to the interactive action information of the target object can be determined, that is, the text content "insert content A in the central area" is converted into an image, text or video of "content A" that can be displayed by the display device. Based on the interaction result, the projection content of the projector can be determined to be "content A" displayed in the central area of ​​the display surface of the display device.

[0055] In an optional embodiment, before acquiring the interaction action information of the target object, the projection content of the projector can be a set projection content, so that the display device used to display the projection content can display the set projection content; after acquiring the unique identification information and interaction action information of the target object, and determining the interaction result based on the unique identification information and interaction action information, the set projection content can be modified, marked, etc., so that the target object can directly interact with the projector.

[0056] Without the need for terminal devices, users can directly create, mark, or modify the projected content using the unique identifier and interactive action information of the target object. This improves the interactivity of the projector and ensures the accuracy of the projected content based on the personalized interactive characteristics of the target object, thereby enhancing the user experience.

[0057] Example 3: Figure 7 is a flowchart of a projector interactive control method provided in Example 3 of the present invention. Based on the above examples, this example describes how to determine the target object from among many actual participating objects. As shown in Figure 7, the method includes: step S210, instantly acquiring image information and / or voice information of each actual participating object.

[0058] It is understood that the projector provided in this embodiment of the invention can be applied to scenarios such as teaching and meetings. When the projector is applied to a teaching scenario, the actual participating objects may include the main presenter and the audience. When the projector is applied to a meeting scenario, the actual participating objects may include all participants. Regardless of the scenario, the actual participating objects may be located at the projector site or at a remote terminal connected to the projector. In this embodiment of the invention, the number and type of actual parameter objects can be determined according to the application scenario of the projector, and this embodiment of the invention does not impose specific limitations on this.

[0059] After the projector starts working, the projector's image acquisition module can acquire image information of each participating object located at the projector site, the camera of the remote terminal connected to the projector can acquire image information of the actual participating objects at the remote terminal, the projector's voice input module can acquire voice information of each participating object at the projector site, and the microphone of the remote terminal connected to the projector can acquire voice information of the actual participating objects at the remote terminal.

[0060] Step S220 determines the action identifier of each actual participating object based on the image information and / or voice information of each actual participating object.

[0061] The action identifiers for actual participating actions can include hand gestures, body movements, facial expressions, vocalizations, or combinations of actions and vocalizations of the participating objects. In one exemplary embodiment, the hand gestures of each participating object can be determined from the collected image information, and these hand gestures can serve as the action identifiers for that object. In another exemplary embodiment, the vocalizations of each participating object can be determined from the acquired vocal information, and these vocalizations can serve as the action identifiers for that object.

[0062] Step S230: Determine the actual participating object whose action is identified by a preset identifier as the target object.

[0063] Specifically, when the action identifier is a gesture, the preset identifier can be a preset gesture, such as a palm facing outward, a fist clenched, or other gestures; when the action identifier is a facial expression, the preset identifier can be a preset facial expression, such as a smile or rapid blinking; when the action identifier is uttering speech, the preset identifier can be preset speech content, such as "Look at me," "Listen to me," or "Let me say something." It should be noted that the preset identifiers and action identifiers correspond to each other, and the specific implementation can be designed according to actual needs; this embodiment of the invention does not impose specific limitations on this.

[0064] In an exemplary embodiment, when the projector is used in a meeting scenario, the actual participants are the actual participating objects. If there are multiple participating objects, during the meeting, the projector's data processing module can acquire image and / or voice information of each participating object. Based on the image and / or voice information of each participating object, it determines the action identifier of each participating object and compares the action identifier of each participating object with a preset identifier. It then determines the action identifier whose difference from the preset identifier is within a preset tolerance range and identifies the participating object executing that action identifier as the current target object. After identifying the target object, the projector's data processing module can acquire image and / or voice information of each participating object, and when the action identifier of another participating object is the preset identifier, it can switch the target object to another participating object with the preset action identifier. In this way, each participating object only needs to perform an action identifier that is the same as or similar to the preset identifier to interact with the projector, enabling the projector to interact with multiple people and improving the convenience of projector interactive control. Then, steps S110 to S130 as shown in Figure 4 can be continued.

[0065] This embodiment acquires image and / or voice information of each actual participating object, and determines the action identifier of each actual participating object based on the image and / or voice information of each actual participating object. The actual participating object with the action identifier of the preset identifier is identified as the target object. In this way, each actual participating object only needs to make the same or similar action identifier as the preset identifier to realize the interaction with the projector, so that the projector can interact with multiple people and improve the convenience of interactive control of the projector.

[0066] Example 4: Figure 8 is a flowchart of a projector interactive control method provided in Example 4 of the present invention. This example describes how to obtain a preset interaction model based on the above examples. As shown in Figure 8, the method includes: step S310, obtaining the unique identifier information and interaction feature information of each initial participating object.

[0067] The initial participants can be the personnel who enter information when using the projector for the first time. The unique identifier information of each initial participant can be determined based on the image and / or voice information of each initial participant; similarly, the interaction feature information of each initial participant can also be determined based on the image and / or voice information of each initial participant. This interaction feature information can include personalized interaction information of the initial participant, and the personalized interaction information of different initial participants will differ.

[0068] Step S320: Based on the unique identifier information and interaction feature information of each initial participating object, construct a model training database in which the unique identifier information and interaction feature information correspond one-to-one.

[0069] After obtaining the unique identifier information and interaction feature information of the initial participating objects, each initial participating object can be encoded, and the unique identifier information and interaction feature information of the same initial participating object can be stored in the encoding position of the initial participating object, thereby constructing a model training database.

[0070] Step S330: Train the deep learning model using the model training database to obtain the preset interactive model.

[0071] Deep learning models, which learn complex features from data through multi-layered neural networks, are a type of machine learning model that can be used to learn the features of interaction information in the model training database. In this embodiment, the unique identifiers and interaction features from the model training database are input into the deep learning model. Based on the output of the deep learning model, the model parameters are adjusted to train the deep learning model and obtain a preset interaction model.

[0072] It is understandable that different initial participating objects have different unique identification information. For different unique identification information and interaction feature information, the model parameters of the deep learning module can be trained accordingly, so that the model parameters of the preset interaction model corresponding to different unique identification information are different. Thus, when the unique identification information and interaction action information of the target object are input into the preset interaction model, the required model parameters of the preset interaction model can be determined first based on the unique identification information of the preset interaction model. Then, the interaction action information is processed based on the determined model parameters of the preset interaction model, and finally, the personalized interaction result corresponding to the interaction action information of the target object is determined, improving the accuracy of the interaction result. Then, steps S110 to S130 as shown in Figure 4 can be continued.

[0073] This embodiment acquires the unique identifiers and interaction features of each initial participating object and constructs a model training database that corresponds one-to-one between the unique identifiers and interaction features. After training a deep learning model using this database to obtain a preset interaction model, the model parameters of each initial participating object can be determined based on its personalized interaction features. Similarly, after identifying a target object, its corresponding model parameters can be determined based on its unique identifiers. By inputting the interaction action information of the target object into the preset interaction model, the personalized interaction result of the target object can be accurately determined, ensuring high accuracy. Consequently, when controlling the projector's projection content based on this interaction result, the projected content also exhibits high accuracy.

[0074] Example 5: Figure 9 is a flowchart of a projector interactive control method provided in Example 5 of the present invention. This example describes how to update the model training database based on the above examples. As shown in Figure 9, the method includes the steps shown in Figure 8: steps S310, S320, and S330. To update the model training database, step S410 is further included between steps S320 and S330: updating the model training database based on the unique identifier information of the target object and the interaction action information.

[0075] It is understandable that the raw data in the model training database includes unique identifier information and interaction feature information determined based on the image information and / or voice information of each initial participating object. This makes the interaction feature information corresponding to the unique identifier information of each initial participating object relatively simple, that is, the raw data in the model training database is relatively simple, which will affect the accuracy of the determined interaction results.

[0076] In this embodiment, during the use of the projector, the target object may have different interactive action information under different projection requirements. At this time, the preset interaction model can continue to learn the interactive action information of the target object, which can make the model parameters in the preset interaction model corresponding to the unique identification information of the target object more accurate.

[0077] Specifically, after obtaining the unique identifier information and interaction action information of the target object, the model training database can be updated accordingly, making the interaction information corresponding to the unique identifier information in the model training database richer and more diverse. Thus, by continuing to train the deep learning model using the updated model training database, the model parameters and their tolerance range can be further adjusted to adapt to different interaction action recognition situations, thereby achieving higher accuracy in determining the interaction results.

[0078] In an optional embodiment, updating the model training database based on the unique identifier information and interaction action information of the target object may include: determining whether the unique identifier information of the target object exists in the model training database; if so, updating the interaction feature information corresponding to the unique identifier information of the target object in the model training database based on the unique identifier information and interaction action information of the target object.

[0079] In another optional embodiment, if the unique identifier information of the target object does not exist in the model training database, then the interaction action information is used as the interaction feature information corresponding to the unique identifier information of the target object; the unique identifier information of the target object and the interaction feature information of the target object are added to the model training database.

[0080] It is understandable that the original data in the model training database consists of unique identifiers and interaction information of the initial participating objects. In the actual use of the projector, the actual participating objects may be the same as or different from the initial participating objects, and the target object is one of the actual participating objects. This allows the model training database to include or exclude the unique identifiers of the target object.

[0081] Specifically, after obtaining the unique identifier information of the target object, this unique identifier information can be compared with existing unique identifier information in the model training database to determine whether the target object's unique identifier information exists in the database. If the target object's unique identifier information exists in the existing database, the target object's interaction action information can be fused with the existing interaction feature information, so that the fused interaction feature information can more accurately represent the target object's personalized characteristics. The identification information can be further encoded based on the original encoding. The unique identification information of the target object is stored at the corresponding encoded address, and the interaction action information of the target object is stored as its interaction feature information at the same encoded address. This allows for the training of a deep learning model based on the unique identification information and interaction feature information of the target object, resulting in model training parameters corresponding to the unique identification information of the target object. Therefore, after inputting the unique identification information and interaction action information of the target object into a preset interaction model, the personalized interaction result of the target object can be determined. Then, steps S110 to S130 as shown in Figure 4 can be continued.

[0082] This embodiment updates the data in the model training database after obtaining the unique identifier information and interaction action information of the target object. This enriches the data in the model training database, so that after training the deep learning model with the updated model training database to obtain the preset interaction model, a more accurate interaction result can be determined based on the preset interaction model. Therefore, when controlling the projection content of the projector according to the interaction result, the projection content has high accuracy.

[0083] Example 6: Figure 10 is a flowchart of a projector interactive control method provided in Example 6 of the present invention. This example describes how to accurately control the projected content based on the above examples. As shown in Figure 10, the method includes: step S510, instantly acquiring image information and / or voice information of the target object.

[0084] Specifically, by acquiring image and / or voice information of the target object in real time, the gestures and / or sounds emitted by the target object can be obtained in real time during the interaction between the target object and the projector. Based on the acquired gestures and / or sounds, the gesture trajectory and voice content of the target object can be determined, and thus the interaction action information of the target object can be determined. Simultaneously, based on the acquired image and voice information of the target object, facial feature information and voiceprint feature information of the target object can be determined, thereby identifying the unique identifier of the target object. Then, steps S110 to S130 as shown in Figure 4 can be continued.

[0085] In step S520, the projection content is adjusted based on the image information and / or voice information of the target object.

[0086] It is understandable that the preset interaction model is based on training a deep learning model, and since a deep learning model is a machine learning model, the preset interaction model can output corresponding interaction results based on the model parameters determined after machine learning. However, since the gestures or sounds made by the target object at different times may vary, the interaction results output by the preset interaction model trained only on historical data may have some deviations, resulting in inaccurate output interaction results, which in turn leads to inaccurate projected content from the projector.

[0087] In this embodiment, after determining the projection content of the projector based on the interaction result output by the preset interaction model, the projector can project the projection content onto the corresponding display device for display. At this time, the target object can determine whether the projection content is what the target object wants to express based on the displayed projection content. For example, if the projection content displayed on the display device is what the target object wants to express, the target object can issue a gesture action or voice content to confirm the projection content; conversely, if the projection content displayed on the display device is not what the target object wants to express, the target object can issue a gesture action or voice content to modify the projection content. Thus, based on the currently acquired image information and / or voice information of the target object, the current gesture action and / or voice content of the target object can be determined. By judging whether the current gesture action of the target object is a gesture action to confirm the projection content, or whether the current voice content of the target object is a voice content to confirm the projection content, it can be determined whether the projection content currently displayed on the display device is what the target object wants to express; if it is determined based on the image information and voice information of the target object that the target object has performed a gesture action to confirm the projection content or issued a voice content to confirm the projection content, the projector can continue to project with this projection content; conversely, if it is determined based on the image information and / or voice information of the target object that the target object has performed a gesture action to modify the projection content or issued a voice content to modify the projection content, the projection content of the projector can be modified according to the current gesture action or voice content of the target object to improve the accuracy of the projection content of the projector.

[0088] In an exemplary embodiment, after inputting the unique identification information and interaction action information of the target object into the preset interaction model, the interaction result output by the preset interaction model is that the target object has written a character '太' (too), but what the target object actually wants to express is the character '大' (big); at this time, after determining that the projection content of the projector is to project the character '太' based on the interaction result and displaying the character '太' on the display device, the target object can issue the voice content 'Modify to the character 大 (big)', and by acquiring the voice information including this voice content, the projection content of the projector can be correspondingly modified from the character '太' to the character '大'.

[0089] In other exemplary embodiments, when the unique identification information and interaction action information of the target object are input into a preset interaction model, and the interaction result output by the preset interaction model is that the target object draws a line segment in the center area of ​​the display surface of the display device, but the target object actually wants to draw a line segment in the upper right corner; at this time, when it is determined from the interaction result that the projection content of the projector is to display a line segment in the center area of ​​the display device, the target object can perform a gesture action to move the line segment in the center area, so as to obtain the image information including the gesture action, and can correspondingly modify the projection content of the projector from displaying a line segment in the center area to displaying a line segment in the upper right corner.

[0090] It should be noted that the above description only illustrates the modification of the projector's projection content by the target object. In the embodiments of the present invention, the method by which the target object modifies the projector's projection content can be designed according to actual needs, and the embodiments of the present invention do not impose specific limitations on this.

[0091] Based on the above embodiments, the projector interactive control method may further include: updating the interactive results output by the preset interactive model according to the adjusted projection content.

[0092] Specifically, during the interaction between the target object and the projector, the data in the model training database can be updated in real time. Based on the updated database, the deep learning model can be trained again, enabling the pre-defined interaction model to learn more personalized feature information. Based on the personalized feature information of each target object, the interaction result corresponding to the interaction action information of that target object can be determined, and different interaction actions of the target object can correspond to different interaction results. However, due to the limitations of the data in the model training database during machine learning, the personalized feature information learned is limited. This leads to a certain deviation between the interaction result output by the pre-defined interaction model and the actual content expressed, resulting in a certain inaccuracy in determining the projected content based on the interaction result. When the content projected by the projector is not what the target object wants to express, the target object can modify the projected content. At this time, the interaction result output by the preset interaction model after inputting the unique identification information and interaction action information of the target object can be updated according to the modified projected content. This ensures that when the target object has the same interaction action information again, the preset interaction model can output an accurate interaction result, thereby improving the accuracy of the interaction between the target object and the projector.

[0093] This embodiment, after controlling the projector's projection content, can adjust the projection content based on real-time acquired image and / or voice information, so that the projector's projection content can accurately express the content that the target object wants to express, thereby further improving the accuracy and convenience of the interaction between the target object and the projector.

[0094] It will be apparent to those skilled in the art to which this disclosure pertains that various modifications and variations can be made to the structure of this disclosure without departing from its scope or spirit. In view of the foregoing, the scope of protection of this disclosure also covers modifications and variations made within the scope of the appended patent applications.

[0095] 10: Data Processing Module 20: Image Acquisition Module 21: Camera 30: Voice Input Module 40: Projection Output Module 50: Wireless communication module 51: Remote Terminal 52: Server 60: Storage Module 100: Projector 101: Outer shell S110~S130: Steps S210~S230: Steps S310~S330: Steps S410: Steps S510~S520: Steps

[0096] Domestic storage information (please note in order of storage institution, date, and number) none Overseas storage information (please note in the order of storage country, institution, date, and number) none

Claims

1. A projector interactive control method, comprising: Real-time acquisition of image and / or voice information of each actual participating object; Based on the image information and / or voice information of each actual participating object, determine an action identifier for each actual participating object; determine the actual participating object whose action identifier is a preset identifier as a target object; obtain a unique identifier and an interactive action information for the target object; input the unique identifier and the interactive action information of the target object into a preset interaction model to obtain an interaction result corresponding to the interactive action information; And based on the interaction result, generate a first projection content and control the projector to project the first projection content, wherein the projector interaction control method further includes: instantly acquiring image information and / or voice information as feedback from the target object to the first projection content; Based on the image information and / or voice information of the target object, a second projection content is generated, and the projector is controlled to adjust the first projection content to the second projection content; and based on the second projection content, the interaction result output by the preset interaction model is updated.

2. The projector interactive control method as described in claim 1 further includes: Obtain a unique identifier and an interaction feature information for each initial participating object; Based on the unique identifier information and interaction feature information of each initial participating object, a model training database is constructed that corresponds one-to-one with the unique identifier information and the interaction feature information; and a deep learning model is trained using the model training database to obtain the preset interaction model.

3. The projector interactive control method as described in claim 2 further includes: The model training database is updated based on the unique identifier information of the target object and the interaction action information.

4. The projector interaction control method as described in claim 3, wherein updating the model training database based on the unique identifier information of the target object and the interaction action information includes: Determine whether the unique identifier information of the target object exists in the model training database; If so, then based on the unique identifier information of the target object and the interaction action information, update the interaction feature information in the model training database corresponding to the unique identifier information of the target object.

5. The projector interactive control method as described in claim 4, wherein updating the model training database based on the unique identifier information of the target object and the interactive action information further includes: If the unique identifier information of the target object does not exist in the model training database, then the interaction action information is used as the interaction feature information corresponding to the unique identifier information of the target object; and the unique identifier information of the target object and the interaction feature information of the target object are added to the model training database.

6. The projector interactive control method as described in claim 1, wherein obtaining the unique identifier information and the interactive action information of the target object includes: Instantly acquire an image and / or a voice message of the target object; And based on the image information and / or voice information of the target object, determine the unique identifier information and the interactive action information of the target object.

7. The projector interactive control method as described in claim 6, wherein determining the interactive action information of the target object based on the image information and / or the voice information of the target object includes: Based on the image information and / or the voice information of the target object, determine an actual action information and an action reference information of the target object; And based on the actual action information and the action reference information, determine the interactive action information of the target object.

8. A projector capable of executing the interactive control method of the projector according to any one of claims 1-7, comprising: An image acquisition module is used to acquire image information; A voice input module is used to obtain voice information; A projection output module is used to output a projection content; and a data processing module is connected to the image acquisition module, the voice input module and the projection output module respectively, for acquiring the image information and the voice information, and controlling the projection output module to output the projection content.

9. The projector as claimed in claim 8, wherein the image acquisition module includes an array of multiple cameras for acquiring image information from different viewpoints around the projector.

10. The projector as described in claim 8, further comprising: A wireless communication module, wherein the data processing module interacts with a remote terminal and / or a server via the wireless communication module.

Images