HRTF implementation method, apparatus and device, and computer readable storage medium

Abstract

The invention discloses an HRTF implementation method, device and equipment and a computer readable storage medium, relates to the technical field of HRTFs, is applied to an earphone terminal, and comprises the steps that a human body area image collected by the earphone terminal is acquired, the human body area image comprises a first human body area image for training and determining a transmission function and a second human body area image for matching and determining the transmission function; performing training according to the first human body region image to obtain first HRTF information; determining second HRTF information according to the second human body region image and a preset transmission function library; and determining a target HRTF parameter according to the first HRTF information and the second HRTF information so as to realize the HRTF based on the target HRTF parameter. According to the method and the device, the implementation accuracy of the HRTF is improved.

Description

Technical Field

[0001] The present application relates to the field of HRTF technology, and in particular to a method, device, equipment and computer-readable storage medium for implementing HRTF. Background Art

[0002] With the development of HRTF (Head-Related Transfer Function), users have also put forward higher requirements for the implementation of HRTF.

[0003] The traditional HRTF implementation method is to take a static photo of the user through the camera and scanning function of an external device, and then directly match similar HRTFs in the existing transfer function library based on the static photo image to implement HRTF. This HRTF implementation method has a great flaw. It can only directly match similar HRTFs in the existing transfer function library (when there are fewer HRTFs in the existing transfer function library), which results in low accuracy of HRTF implementation.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a HRTF implementation method, device, equipment and computer-readable storage medium, aiming to solve the technical problem of low accuracy of HRTF implementation.

[0006] To achieve the above object, the present application provides a method for implementing HRTF, which is applied to a headphone terminal and includes:

[0007] Acquire a human body region image collected by the headset terminal, wherein the human body region image includes a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function;

[0008] Performing training according to the first human body region image to obtain first HRTF information;

[0009] Determining second HRTF information according to the second human body region image and a preset transfer function library;

[0010] A target HRTF parameter is determined according to the first HRTF information and the second HRTF information to implement the HRTF based on the target HRTF parameter.

[0011] In one embodiment, the first human body region image includes a head image, an auricle image, and an ear canal image, and the step of obtaining the first HRTF information through training according to the first human body region image includes:

[0012] Determine head physiological parameters in the head image, and obtain first HRTF parameters by training in a preset head training model based on the head physiological parameters;

[0013] Determine the auricle physiological parameters in the auricle image, and obtain second HRTF parameters by training in a preset auricle training model based on the auricle physiological parameters;

[0014] Determine the ear canal physiological parameters in the ear canal image, and obtain third HRTF parameters by training in a preset ear canal training model based on the ear canal physiological parameters, and use the first HRTF parameters, the second HRTF parameters and the third HRTF parameters as first HRTF information.

[0015] In one embodiment, after the step of obtaining the first HRTF information through training according to the first human body region image, the following steps are included:

[0016] The first HRTF parameter, the second HRTF parameter and the third HRTF parameter are updated to a preset transfer function library.

[0017] In one embodiment, the second human body region image includes a torso image, and the step of determining the second HRTF information according to the second human body region image and a preset transfer function library includes:

[0018] Determining a fourth HRTF parameter that matches the torso image in a preset transfer function library, and detecting whether a first matching value of the fourth HRTF parameter is greater than a preset first threshold, wherein the first matching value includes a matching degree between the torso image and the fourth HRTF parameter;

[0019] When the first matching value of the fourth HRTF parameter is greater than a preset first threshold, the fourth HRTF parameter is used as the second HRTF information.

[0020] In one embodiment, before the step of obtaining the first HRTF information through training according to the first human body region image, the step includes:

[0021] Based on the head image, the auricle image and the ear canal image in the first human body region image, the matching target HRTF parameters are determined in a preset transfer function library in turn, and a second matching value of the target HRTF parameters is detected to be greater than a preset second threshold, wherein the second matching value includes a matching degree between the head image, the auricle image or the ear canal image and the target HRTF parameters, and the second matching value is greater than the first matching value;

[0022] When the second matching value of the target HRTF parameter is greater than a preset second threshold, using the target HRTF parameter as the second HRTF information;

[0023] When the second matching value of the target HRTF parameter is less than or equal to a preset second threshold, the step of obtaining the first HRTF information through training according to the first human body region image is performed.

[0024] In one embodiment, the first HRTF information includes first HRTF parameters of the head image, second HRTF parameters of the auricle image, and third HRTF parameters of the ear canal image, and the second HRTF information includes fourth HRTF parameters of the torso image. The step of determining the target HRTF parameters according to the first HRTF information and the second HRTF information includes:

[0025] A convolution sum among the first HRTF parameter, the second HRTF parameter, the third HRTF parameter and the fourth HRTF parameter is determined, and the convolution sum is used as a target HRTF parameter.

[0026] In one embodiment, the headset terminal includes a camera and a position sensor, and the step of acquiring the human body region image captured by the headset terminal includes:

[0027] Acquire the wearing process image captured by the camera and the position state captured by the position sensor in a preset first state, wherein the first state includes that the connection state of the headset terminal is the first connection and / or the input control instruction matches the preset acquisition instruction;

[0028] The wearing process image is divided based on the position state and preset training features to obtain a human body region image.

[0029] In one embodiment, the HRTF implementation method further includes:

[0030] Acquiring input physiological parameters, wherein the physiological parameters include trunk physiological parameters, head physiological parameters, auricle physiological parameters and ear canal physiological parameters;

[0031] Determine corresponding torso HRTF parameters based on the torso constructed physiological parameters, and determine corresponding head HRTF parameters based on the head constructed physiological parameters;

[0032] Based on the physiological parameters of the auricle, the corresponding auricle HRTF parameters are determined; based on the physiological parameters of the ear canal, the corresponding ear canal HRTF parameters are determined; and the torso HRTF parameters, the head HRTF parameters, the auricle HRTF parameters and the ear canal HRTF parameters are summarized as a preset transfer function library.

[0033] In addition, to achieve the above purpose, the present application also provides a HRTF implementation device, which is applied to a headphone terminal, and the HRTF implementation device includes:

[0034] An image acquisition module, used to acquire a human body region image collected by the headset terminal, wherein the human body region image includes a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function;

[0035] A first processing module, configured to obtain first HRTF information by training according to the first human body region image;

[0036] a second processing module, configured to determine second HRTF information according to the second human body region image and a preset transfer function library;

[0037] A function determination module is used to determine target HRTF parameters according to the first HRTF information and the second HRTF information, so as to implement HRTF based on the target HRTF parameters.

[0038] In addition, to achieve the above-mentioned purpose, the present application also provides an HRTF implementation device, including a camera, a processor, a memory, and an HRTF implementation method program stored in the memory and executable by the processor, wherein when the HRTF implementation method program is executed by the processor, the steps of the HRTF implementation method described above are implemented.

[0039] The present application also provides a computer-readable storage medium, on which a HRTF implementation method program is stored, wherein when the HRTF implementation method program is executed by a processor, the steps of the HRTF implementation method described above are implemented.

[0040] An embodiment of the present application provides a method for implementing HRTF, which is applied to a headset terminal, by acquiring a human body area image captured by the headset terminal, wherein the human body area image includes a first human body area image for training to determine a transfer function and a second human body area image for matching to determine a transfer function; training is performed according to the first human body area image to obtain first HRTF information; second HRTF information is determined according to the second human body area image and a preset transfer function library; target HRTF parameters are determined according to the first HRTF information and the second HRTF information, so as to implement HRTF based on the target HRTF parameters, by dividing the human body area image into the first human body area image for training to determine a transfer function and the second human body area image for matching to determine a transfer function, and then respectively executing the processes of training according to the first human body area image to obtain the first HRTF information; determining the second HRTF information according to the second human body area image and a preset transfer function library, and finally determining the target HRTF parameters based on the first HRTF information and the second HRTF information to implement HRTF, thereby avoiding the phenomenon that similar HRTFs can only be directly matched in an existing transfer function library (when there are fewer existing HRTFs in the existing transfer function library), which causes the accuracy of HRTF implementation. This HRTF implementation method not only provides a new HRTF implementation method, but also divides the human body area image into a first human body area image for training to determine the transfer function and a second human body area image for matching to determine the transfer function, and performs matching and training processing respectively to obtain the first HRTF information and the second HRTF information, so as to implement HRTF based on the first HRTF information and the second HRTF information, thereby avoiding the problem of low accuracy caused by the small number of existing HRTFs in the existing transfer function library, thereby improving the accuracy of HRTF implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flow chart of the first embodiment of the HRTF implementation method of the present application;

[0042] Figure 2 A physical schematic diagram of an earphone terminal in the HRTF implementation method of the present application;

[0043] Figure 3 A schematic diagram of a wearing process in the HRTF implementation method of the present application;

[0044] Figure 4 A flowchart of the method for implementing HRTF of the present application is shown;

[0045] Figure 5 This is another flowchart of the HRTF implementation method of the present application;

[0046] Figure 6 This is a schematic diagram of the modules of the device for implementing HRTF of this application;

[0047] Figure 7 This is a schematic diagram of the device structure of the hardware operating environment involved in this application.

[0048] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings.

[0049] Description of Figure Numbers:

[0050] 100. earphone terminal; 110. camera; 120. position sensor; 1001. processing device; 1002. ROM; 1003. storage device; 1004. RAM; 1005. bus; 1006. I / O port; 1007. input device; 1008. output device; 1009. communication device; 200. mobile phone. DETAILED DESCRIPTION

[0051] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0052] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0053] The commonly used HRTF implementation method is to use the camera and scanning function of an external device to take a static photo of the user, and then directly match similar HRTFs in the existing transfer function library based on the static photo image to implement HRTF. However, the transfer function library has a small amount of data, and the similar HRTFs found have a poor match degree and low personalization. On the other hand, it is to directly conduct deep learning on all images, which requires a lot of computing power to achieve HRTF adaptation, so either the accuracy of HRTF implementation is not high or the cost of implementing HRTF is high.

[0054] Therefore, based on the insufficiency of the above HRTF implementation scheme, the HRTF implementation method of the present application is proposed. The solution of the embodiment of the present application is: by dividing the human body region image into a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function, and then respectively executing the process of training according to the first human body region image to obtain the first HRTF information; determining the second HRTF information according to the second human body region image and the preset transfer function library, and finally determining the target HRTF parameters based on the first HRTF information and the second HRTF information to implement the HRTF, so as to avoid the phenomenon that only similar HRTFs can be directly matched in the existing transfer function library (when there are fewer HRTFs in the existing transfer function library), which causes the accuracy of HRTF implementation. This HRTF implementation method not only provides a new HRTF implementation method, but also divides the human body region image into a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function, and respectively performs matching and training processing to obtain the first HRTF information and the second HRTF information, so as to implement the HRTF based on the first HRTF information and the second HRTF information, and avoid the problem of low accuracy caused by the small number of existing HRTFs in the existing transfer function library, thereby improving the accuracy of HRTF implementation.

[0055] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or a device capable of realizing the above functions, an HRTF implementation device, etc. The following takes the HRTF implementation device as an example to illustrate this embodiment and the following embodiments.

[0056] Based on this, the present application embodiment provides a method for implementing HRTF, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the HRTF implementation method of the present application.

[0057] Reference Figure 1 The present application provides a method for implementing HRTF, which is applied to a headphone terminal. The method for implementing HRTF includes:

[0058] Step S10, acquiring a human body region image collected by the headset terminal, wherein the human body region image includes a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function;

[0059] For example, see Figure 2 , Figure 2This is a physical schematic diagram of the headphone terminal in the implementation method of HRTF of the present application. The headphone terminal 100 in the figure is equipped with two or more depth cameras 110 (that is, the image information collected at this time is depth image information) and a position sensor 120. The position sensor 120 can be an optical sensor, a capacitive sensor or a motion sensor (such as an accelerometer / gyroscope), thereby realizing image acquisition during the entire wearing process. It is worth noting that, on the one hand, the headphone terminal 100 can be a component hardware for realizing HRTF, thereby realizing image capture without the use of external mobile phones and other instruments. On the other hand, the controller in the headphone terminal 100 can be the executive device of this embodiment, or other control devices connected to the controller in the headphone terminal 100 as the executive device of this embodiment, which is not limited here. Further, refer to Figure 3 , Figure 3 This is a schematic diagram of a scene of the wearing process in the HRTF implementation method of the present application. The figure takes a Bluetooth headset as an example (of course, it can also be a general smart headset, which is not limited here). The figure sequentially illustrates the process of unpacking the headset terminal 100 and connecting it to the mobile phone 200 through Bluetooth, and then wearing it. At this time, the camera 110 set on the headset terminal 100 can capture the image of the wearing process, which is convenient for subsequent image processing.

[0060] In this embodiment, when the headset terminal equipped with a camera is in the wearing process, the camera on the headset terminal will capture the wearing process image of the headset terminal during the wearing process, and then the human body region image is obtained after the image information is divided and processed. The human body region image includes a first human body region image for training to determine the transmission function and a second human body region image for matching to determine the transmission function. The first human body region image refers to an image that needs to be trained and determined, such as an ear canal image, auricle image, and head image whose human body region image has a large correlation with HRTF and has large differences between different users. It is necessary to train to obtain personalized HRTF parameters. The second human body region image refers to an image that does not need to be trained and determined, and the relevant HRTF parameters are directly determined in the transmission function library. For example, a torso image whose human body region image has a small correlation with HRTF and has small differences between different users can directly match to obtain personalized HRTF parameters. Among them, the wearing process image refers to the image in the process of wearing the headset, and the human body region image refers to the image of different regions of the human body obtained by division. At this time, it can be intelligently processed based on the image captured by the camera on the headset terminal to improve the intelligence of the entire HRTF implementation. It is worth noting that at this time, it is necessary to determine that the image at this time is the wearing process image. The image collected after the Bluetooth connection is sensed can be used as the wearing process image, or after the Bluetooth connection, there is a control instruction to collect the image and the subsequent image can be used as the wearing process image. Figure 4 , Figure 4This is a flow chart of the implementation method of HRTF of the present application. After the earphone terminal ends the Bluetooth connection, it starts to execute the process of B10 to pick up features. At this time, the user's physiological features are picked up. The picking process is the process image of the user moving the earphone from the front to the ear canal. After the wearing is completed at B20, the final picture picking of B30 will be realized. At this time, the final picture is the process picture from the earphone out of the box to the ear. The time from picking up to wearing is short, so a higher refresh rate is required to pick up and save the user graphics, so as to avoid the common method of using a mobile phone to shoot the user's 3D information, which requires the user to be in a stationary state. It is worth noting that because the earphone terminal is equipped with a depth camera, the user does not need to remain stationary, and the user's earphone wearing process is the process of real-time 3D supplementation of the two earphone terminals.

[0061] Step S20, training according to the first human body region image to obtain first HRTF information;

[0062] Step S30, determining second HRTF information according to the second human body region image and a preset transfer function library;

[0063] In this embodiment, when determining the first human body region image and the second human body region image distinguished based on the human body region image, the first human body region image and the second human body region image will be processed respectively, and then the first HRTF information will be obtained by training based on the first human body region image, and the second HRTF information will be determined based on the second human body region image and the preset transfer function library, that is, on the one hand, the relevant image features of the first human body region image are put into the relevant model or algorithm for training to obtain the corresponding HRTF parameters as the first HRTF information; on the other hand, the HRTF parameters whose matching degree reaches the threshold with the second human body region image are directly matched in the preset transfer function library as the second HRTF information, and the preset transfer function library refers to a database storing HRTF parameters of different image regions. It is worth noting that if the preset transfer function library does not have the HRTF parameters matched by the second human body region image, it can also be trained and the HRTF parameters obtained by training are stored in the transfer function library for direct use next time. At this time, the human body region image is divided into the first human body region image and the second human body region image for processing, so as to avoid the phenomenon of directly using the HRTF parameters with the highest matching value when the matching value is not high, so as to determine the best HRTF parameters through training to improve the accuracy of HRTF implementation.

[0064] In one embodiment, referring to Figure 4After obtaining the first human body region image and the second human body region image, the process B40 will be executed to generate corresponding physiological parameters based on the images of different regions, and the process B50 will be executed to extract corresponding HRTF parameters based on the relationship between the characteristics of the physiological parameters and HRTF, thereby realizing the targeted determination of HRTF parameters for images of different regions, that is, personalized HRTF generation, to achieve the effect of personalized HRTF. Further, referring to Figure 5 , Figure 5 This is another flow chart of the HRTF implementation method of the present application, that is, after the generation parameters corresponding to the picture / image are generated, the torso parameters in the picture / image will be matched with the HRTF parameters in the transfer function library, and at the same time other parameters in the picture / image, such as the head, auricle and ear canal, will be trained to obtain the HRTF parameters. Finally, the final HRTF parameters can be determined based on the two HRTF parameters to achieve the HRTF effect based on the final HRTF parameters.

[0065] Step S40: determining target HRTF parameters according to the first HRTF information and the second HRTF information, so as to implement HRTF based on the target HRTF parameters.

[0066] In this embodiment, after determining the first HRTF information and the second HRTF information corresponding to the first human body area image and the second human body area image respectively, the target HRTF parameters will be determined based on the first HRTF information and the second HRTF information, and finally the HRTF can be implemented based on the target HRTF parameters, such as the setting of parameters such as the interaural time difference, interaural sound level difference, scattering of sound waves by the auricle and the reflected sound of the trunk in the target HRTF parameters to control the implementation effect of the HRTF. At this time, the parameters of the reflected sound of the trunk can be parameters determined by the trunk image, the scattering of sound waves by the auricle can be parameters determined by the auricle image, the interaural time difference and the interaural sound level difference can be parameters jointly determined by the ear canal image, the auricle image and the head image. Taking the interaural time difference as a numerical value as an example, the ear canal image can account for 40% of the numerical value, the auricle image can account for 40% of the numerical value, and the head image can account for 20% of the numerical value. For example, the parameter determined by the ear canal image is A, the parameter determined by the auricle image is B and the parameter determined by the head image is C, then the final interaural time difference is 40%A+40%B+20%C. The HRTF parameters in the first HRTF information and the second HRTF information can be convolved and processed to obtain the target HRTF parameters, or other methods can be used, which are not limited here. Because the personalization of the key parameters in the first HRTF information is guaranteed (different users obtain different first HRTF information), the accuracy of the final target HRTF parameters can be guaranteed to achieve accurate HRTF.

[0067] In the present embodiment, a method for realizing HRTF is provided, which is applied to a headset terminal, by acquiring a human body area image captured by the headset terminal, wherein the human body area image includes a first human body area image for training to determine a transfer function and a second human body area image for matching to determine a transfer function; training is performed according to the first human body area image to obtain first HRTF information; second HRTF information is determined according to the second human body area image and a preset transfer function library; target HRTF parameters are determined according to the first HRTF information and the second HRTF information, so as to realize HRTF based on the target HRTF parameters, by dividing the human body area image into the first human body area image for training to determine a transfer function and the second human body area image for matching to determine a transfer function, and then respectively executing the processes of training according to the first human body area image to obtain the first HRTF information; determining the second HRTF information according to the second human body area image and a preset transfer function library, and finally determining the target HRTF parameters based on the first HRTF information and the second HRTF information, so as to realize HRTF, thereby avoiding the phenomenon that similar HRTF can only be directly matched in an existing transfer function library (when there are fewer existing HRTFs in the existing transfer function library), which causes the accuracy of HRTF realization. This HRTF implementation method not only provides a new HRTF implementation method, but also divides the human body area image into a first human body area image for training to determine the transfer function and a second human body area image for matching to determine the transfer function, and performs matching and training processing respectively to obtain the first HRTF information and the second HRTF information, so as to implement HRTF based on the first HRTF information and the second HRTF information, thereby avoiding the problem of low accuracy caused by the small number of existing HRTFs in the existing transfer function library, thereby improving the accuracy of HRTF implementation.

[0068] Further, based on the first embodiment of the present application, a second embodiment of the method for implementing HRTF of the present application is proposed. In this embodiment, in the above step S20, the first human body region image includes a head image, an auricle image, and an ear canal image, and the step of training according to the first human body region image to obtain the first HRTF information includes:

[0069] Step S21, determining head physiological parameters in the head image, and obtaining first HRTF parameters by training in a preset head training model based on the head physiological parameters;

[0070] Step S22, determining the auricle physiological parameters in the auricle image, and obtaining second HRTF parameters by training in a preset auricle training model based on the auricle physiological parameters;

[0071] Step S23, determining the ear canal physiological parameters in the ear canal image, and obtaining third HRTF parameters by training in a preset ear canal training model based on the ear canal physiological parameters, and using the first HRTF parameters, the second HRTF parameters and the third HRTF parameters as the first HRTF information.

[0072] In this embodiment, in order to ensure the personalization of HRTF parameters, it is necessary to train the head image, auricle image and ear canal image in the first human body region image to obtain the corresponding HRTF parameters, thereby ensuring the personalization of HRTF parameters for the head image, auricle image and ear canal image, so as to ensure the personalization and accuracy of HRTF. At this time, the first HRTF parameters are obtained by determining the head physiological parameters in the head image, and training in a preset head training model based on the head physiological parameters. The head training model refers to a model for training the head to obtain HRTF parameters. The first HRTF parameters refer to the parameters obtained by training the head physiological parameters in the head training model. The head physiological parameters refer to the relevant parameters of the head, such as head size: such as the length, width and height of the head; facial size: such as the distance from the eyes to the ears. At the same time, the auricle physiological parameters in the auricle image will be determined, and the second HRTF parameters will be obtained by training in the preset auricle training model based on the auricle physiological parameters. The auricle training model refers to a model for training the auricle to obtain HRTF parameters. The second HRTF parameters refer to the parameters obtained by training the auricle physiological parameters in the auricle training model. The auricle physiological parameters refer to the relevant parameters of the ear, such as the auricle size: including the distance from the auricle to the head, the length and width of the auricle; the inter-ear distance: the distance between the two ears, which is a key factor affecting the time difference of the sound reaching each ear. The ear canal physiological parameters in the ear canal image will also be determined, and the third HRTF parameters will be obtained by training in the preset ear canal training model based on the ear canal physiological parameters. The ear canal training model refers to a model for training the ear canal to obtain HRTF parameters. The third HRTF parameters refer to the parameters obtained by training the ear canal physiological parameters in the ear canal training model. The ear canal physiological parameters refer to the relevant parameters of the ear canal, such as the tortuosity in the ear canal. It is worth noting that the training method for different physiological parameters can be the same as the commonly used physiological parameter training method, which is not limited here. At this time, personalized training is performed on the regional images that have a greater impact on the realization of HRTF, which can improve the accuracy of realizing HRTF. At the same time, the image is divided into multiple regional images for processing, which can greatly reduce the complexity of processing.

[0073] Further, after the step of obtaining the first HRTF information through training according to the first human body region image, the method further includes:

[0074] Step S24: updating the first HRTF parameter, the second HRTF parameter and the third HRTF parameter to a preset transfer function library.

[0075] In this embodiment, after the first HRTF parameters, the second HRTF parameters and the third HRTF parameters are determined, the first HRTF parameters, the second HRTF parameters and the third HRTF parameters will be updated to the preset transfer function library, so that when the preset transfer function library has sufficient data later, the preset transfer function library can be directly used to match the HRTF parameters corresponding to the head image, the auricle image and the ear canal image, thereby improving the efficiency of implementing HRTF.

[0076] Further, based on the first embodiment and / or the second embodiment of the present application, a third embodiment of the method for implementing HRTF of the present application is proposed. In this embodiment, in the above step S30, the second human body region image includes a torso image, and the step of determining the second HRTF information according to the second human body region image and a preset transfer function library includes:

[0077] Step S31, determining a fourth HRTF parameter matching the torso image in a preset transfer function library, and detecting whether a first matching value of the fourth HRTF parameter is greater than a preset first threshold, wherein the first matching value includes a matching degree between the torso image and the fourth HRTF parameter;

[0078] Step S32: when the first matching value of the fourth HRTF parameter is greater than the preset first threshold, the fourth HRTF parameter is used as the second HRTF information.

[0079] In this embodiment, when it is a torso image, because the torso size is easily affected by the size and style of clothes, the torso image can be matched to the HRTF parameters in the preset transfer function library, and the torso image has little influence on the realization of HRTF. Therefore, the fourth HRTF parameter matching the torso image can be determined in the preset transfer function library, and it is detected whether the first matching value of the fourth HRTF parameter is greater than the preset first threshold, wherein the first matching value includes the matching degree between the torso image and the fourth HRTF parameter, and then when the first matching value of the fourth HRTF parameter is greater than the preset first threshold, the fourth HRTF parameter is used as the second HRTF information. Among them, the fourth HRTF parameter refers to all HRTF parameters that match the torso image in the transfer function library, the first matching value refers to the matching value of each HRTF parameter and the torso image, and the preset first threshold refers to the matching value defined by the user, so that when the first matching value of the fourth HRTF parameter is greater than the preset first threshold, it can be determined that the HRTF parameters that meet the requirements are determined in the transfer function library, and the matching degree between the HRTF parameters and the torso image can be calculated based on the existing matching degree calculation method, such as determining the matching degree between the HRTF parameters and each feature in the torso image respectively, and then determining the average value of all feature matching degrees as the first matching value, so that the second HRTF information can be quickly determined in the transfer function library to improve the efficiency of HRTF implementation. It is worth noting that if there is no fourth HRTF parameter matching the torso image in the transfer function library, it can also be determined by algorithm training.

[0080] Furthermore, before the step of obtaining the first HRTF information through training according to the first human body region image, the method further includes:

[0081] Step S201, based on the head image, the auricle image and the ear canal image in the first human body region image, determining the matching target HRTF parameters in a preset transfer function library in turn, and detecting whether a second matching value of the target HRTF parameters is greater than a preset second threshold, wherein the second matching value includes a matching degree between the head image, the auricle image or the ear canal image and the target HRTF parameters, and the second matching value is greater than the first matching value;

[0082] Step S202, when the second matching value of the target HRTF parameter is greater than a preset second threshold, using the target HRTF parameter as the second HRTF information;

[0083] Step S203, when the second matching value of the target HRTF parameter is less than or equal to the preset second threshold, executing the step of obtaining first HRTF information through training according to the first human body region image.

[0084] In this embodiment, before determining the first HRTF information, it is necessary to determine whether there are HRTF parameters matching the first human body region image in the preset transfer function library. Based on the head image, auricle image and ear canal image in the first human body region image, the matching target HRTF parameters are determined in the preset transfer function library in turn. Then, by detecting whether the second matching value of the target HRTF parameter is greater than the preset second threshold, it can be determined whether the HRTF parameter can be directly determined in the transfer function library or whether the HRTF parameter needs to be determined based on training. That is, as long as the previous training reaches a certain data richness, it can be directly determined in the transfer function library. That is, when the second matching value of the target HRTF parameter is greater than the preset second threshold, the target HRTF parameter is used as the second HRTF information. Otherwise, the step of obtaining the first HRTF information through training according to the first human body region image is continued, and then the HRTF parameter is determined through training. The target HRTF parameter refers to the HRTF parameter matching the head image, auricle image or ear canal image. The second matching value is greater than the first matching value mainly because the head image, auricle image and ear canal image have a greater impact on the effect of HRTF than the torso image, so a higher matching degree needs to be set. It is worth noting that the target HRTF parameters corresponding to at least one of the head image, auricle image or ear canal image can be determined, and the other target HRTF parameters that cannot be determined can be determined by training, or all of them can be determined by training or matching, which is not limited here.

[0085] Further, based on the first embodiment, the second embodiment and / or the third embodiment of the present application, a fourth embodiment of the HRTF implementation method of the present application is proposed. In this embodiment, in the above step S40, the first HRTF information includes the first HRTF parameters of the head image, the second HRTF parameters of the auricle image and the third HRTF parameters of the ear canal image, the first HRTF information includes the fourth HRTF parameters of the torso image, and the step of determining the target HRTF parameters according to the first HRTF information and the second HRTF information includes:

[0086] Step S41, determining the convolution sum among the first HRTF parameter, the second HRTF parameter, the third HRTF parameter and the fourth HRTF parameter, and taking the convolution sum as the target HRTF parameter.

[0087] In this embodiment, the human body area image includes a torso image, a head image, an auricle image and an ear canal image, and then a personalized HRTF is performed based on the torso image, the head image, the auricle image and the ear canal image. On the one hand, the torso image is directly matched with the relevant fourth HRTF parameters, and at the same time, combined with model algorithms such as deep learning, the first HRTF parameters corresponding to the head image, the second HRTF parameters corresponding to the auricle image and the third HRTF parameters corresponding to the ear canal image are determined respectively, so as to realize truly personalized HRTF and achieve more natural spatial audio rendering. Final picture pickup: After wearing, the final state image is picked up on the L / R headphones. Among them, the fourth HRTF parameter is the HRTF parameter corresponding to the torso image, which can generally be matched in the HRTF parameter library, the first HRTF parameter is the HRTF parameter corresponding to the head image, which generally needs to be trained to obtain personalized parameters, the second HRTF parameter is the HRTF parameter corresponding to the auricle image, which generally needs to be trained to obtain personalized parameters, and the third HRTF parameter is the HRTF parameter corresponding to the ear canal image, which generally needs to be trained to obtain personalized parameters. Finally, the target HRTF parameter can be determined based on the first HRTF parameter, the second HRTF parameter, the third HRTF parameter and the fourth HRTF parameter, and the determination method can be to average the four parameters, or to convolve the four parameters. Because the HRTF parameters corresponding to the head image, the auricle image and the ear canal image have a small matching relationship with each other, and are directly related to the final effect, the HRTF parameters corresponding to the head image, the auricle image and the ear canal image are trained to ensure personalized HRTF implementation, and at the same time, the acquisition is performed based on the acquisition method of the headset terminal, and the HRTF parameters are determined by intelligently dividing into multiple areas, respectively, to ensure the intelligence and accuracy of the entire HRTF implementation.

[0088] In one embodiment, the HRTF implementation method further includes:

[0089] Step S40, obtaining input physiological parameters, wherein the physiological parameters include trunk physiological parameters, head physiological parameters, auricle physiological parameters and ear canal physiological parameters;

[0090] Step S50, determining corresponding torso HRTF parameters based on the torso constructed physiological parameters, and determining corresponding head HRTF parameters based on the head constructed physiological parameters;

[0091] Step S60, determine the corresponding auricle HRTF parameters based on the auricle physiological parameters, determine the corresponding ear canal HRTF parameters based on the ear canal physiological parameters, and summarize the torso HRTF parameters, head HRTF parameters, auricle HRTF parameters and ear canal HRTF parameters as a preset transfer function library.

[0092] In this embodiment, before executing the entire HRTF implementation method, the transmission function library will be constructed, that is, the physiological parameters are constructed and input, and then the transmission function library is constructed based on the constructed physiological parameters. Among them, the constructed physiological parameters refer to the physiological parameters for constructing the HRTF parameters in the transmission function library, including the trunk constructed physiological parameters for constructing the trunk HRTF parameters, such as the trunk length, width, etc.; the head constructed physiological parameters for constructing the head HRTF parameters, such as the head shape, plane features, etc.; the auricle constructed physiological parameters for constructing the auricle HRTF parameters, such as the size and shape of the ear; the ear canal constructed physiological parameters for constructing the ear canal HRTF parameters, such as the shape of the ear, the concave and convex situation, etc. Then, the corresponding HRTF parameters can be constructed based on the above constructed physiological parameters, and the construction process can be obtained by training in a specific training model based on the constructed physiological parameters of each position, or other methods, which are not limited here. Finally, the preset transmission function library can be obtained by summarizing the trunk HRTF parameters, the head HRTF parameters, the auricle HRTF parameters and the ear canal HRTF parameters, so that the parameters can be directly matched in the preset transmission function library later, and then the personalized HRTF can be quickly realized based on different user characteristics.

[0093] In one embodiment, after determining the human body region image, physiological parameters will be determined based on the three-dimensional data of the human body region image, including size information (length, width, height, ear size and shape, etc.), geometric parameters (curvature, angle, area, volume, etc. of the surface), and morphological parameters (such as the shape characteristics of the auricle, head, and ear canal, etc.). Finally, it can be extracted based on the determined physiological parameters and HRTF relationship: this function is based on a huge personalized HRTF library (i.e., a preset transfer function library, which consists of four parts, the trunk class Ha, the head class Hb, the auricle class Hc and the ear canal class Hd, then a user's HRTF is H = Ha * Hb * Hc * Hd). In particular, in order to reduce computing power, first search for completely consistent parameter matches in the trunk, head, ear canal and auricle categories. Often the trunk part can match the completely consistent Has (because the trunk size is easily affected by the size and style of clothes, at this time Has represents the parameters matched by the user in the Ha library), while Hb, Hc and Hd are almost difficult to match due to the current individual uniqueness. In order to increase the accuracy of personalized HRTF, Hb, Hc and Hd are recreated in this dimension through deep learning to obtain Hbn, Hcn and Hdn for the user. At the same time, Hbn, Hcn and Hdn are uploaded to the cloud library to supplement the HRTF library. Finally, personalized HRTF generation is achieved: at this time, the HRTF of the user is H = Has*Hbn*Hcn*Hdn. Finally, personalized HRTF can be achieved based on H to ensure the implementation effect of HRTF.

[0094] Further, based on the first embodiment, the second embodiment, the third embodiment and / or the fourth embodiment of the present application, a fifth embodiment of the implementation method of the HRTF of the present application is proposed. In this embodiment, the headset terminal includes a camera and a position sensor, and the step of obtaining the human body area image collected by the headset terminal includes:

[0095] Step a, acquiring a wearing process image captured by a camera and a position state captured by a position sensor in a preset first state, wherein the first state includes that the connection state of the headset terminal is the first connection and / or the input control instruction matches the preset acquisition instruction;

[0096] Step b: dividing the wearing process image based on the position state and preset training features to obtain a human body area image.

[0097] In this embodiment, when the human body area image is obtained, the current state information of the headset terminal will be obtained first, wherein the current state information includes the connection state and control instructions of the headset terminal, and then based on the current state information, it is determined whether the image collected at this time needs to be used as the human body area image. The connection state of the headset terminal refers to the Bluetooth connection state of the headset terminal, such as connected, first connection, etc. The control instruction refers to the control instruction input to the headset terminal, such as control shutdown, control image acquisition, etc. Then, when the connection state is the first connection, the wearing process image of the headset terminal during the wearing process is collected based on the camera, and / or, when the connection state is not the first connection, and the control instruction matches the preset acquisition instruction, the wearing process image of the headset terminal during the wearing process is collected based on the camera, that is, there is a need for image acquisition or the first Bluetooth connection is met in at least one of the situations, and the image at this time can be obtained. It is worth noting that at this time, the sensor in the headset terminal can be used to detect whether the headset terminal is in a mobile state, so as to wake up the camera when it is in a mobile state to collect the wearing process image of the headset terminal during the wearing process. For example, when the headset terminal is used for the first time, the headset terminal is generally placed in the headset box to connect to Bluetooth, and then worn after the Bluetooth connection is completed. Therefore, after the first Bluetooth connection, the personalized wearing process image acquisition can be automatically entered; if the current state of the headset terminal and the mobile phone have been paired with Bluetooth, you can choose whether to perform personalized wearing process image acquisition on the interface. If personalized acquisition is not performed (that is, after executing steps S10-S30 to realize image acquisition, the new target HRTF parameters are re-determined for HRTF control), the system will use the personalized wearing process image information collected by the original wearer. If personalized wearing process image acquisition is selected, the headset terminal needs to be put back from the headset to the headset box, and the headset terminal is taken out to start personalized wearing process image acquisition. It is worth noting that the headset terminal can retain multiple personalized wearing process images, or only one personalized data, so that the historical personalized wearing process images can be quickly used, thereby greatly improving the implementation efficiency of HRTF. Furthermore, after the wearing process image is collected, the wearing process image will be divided based on the position state and the preset training features to obtain the human body region image, that is, to determine which frame of the wearing process image belongs to which region of the image, the position state refers to the position state collected by the position sensor, generally the position state of the ear canal, and the preset training features refer to the training features of different regions, such as the training features of the trunk, head and auricle, and then matched with the wearing process image based on the training features. Images of different positions are obtained, that is, human body region images. It is worth noting that at this time, the ear canal image can be directly determined based on the position state, the trunk, head and auricle images can be determined based on the training features, or the trunk and head can be determined based on the training features, and finally the auricle image can be determined based on the determined ear canal and head. The determination process of the human body region image is not limited.

[0098] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the implementation method of the HRTF of the present application. More simple transformations based on this technical concept are all within the protection scope of the present application.

[0099] This application also provides a HRTF implementation device, please refer to Figure 4 , the HRTF implementation device is applied to the headphone terminal, and the HRTF implementation device includes:

[0100] An image acquisition module A10 is used to acquire a human body region image collected by the headset terminal, wherein the human body region image includes a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function;

[0101] A first processing module A20, configured to obtain first HRTF information by training according to the first human body region image;

[0102] A second processing module A30, configured to determine second HRTF information according to the second human body region image and a preset transfer function library;

[0103] The function determination module A40 is used to determine target HRTF parameters according to the first HRTF information and the second HRTF information, so as to implement HRTF based on the target HRTF parameters.

[0104] The HRTF implementation device provided in the present application adopts the HRTF implementation method in the above-mentioned embodiment, which can solve the technical problem of low accuracy of HRTF implementation. Compared with the prior art, the beneficial effects of the HRTF implementation device provided in the present application are the same as the beneficial effects of the HRTF implementation method provided in the above-mentioned embodiment, and other technical features in the HRTF implementation device are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0105] The present application provides a HRTF implementation device (which can be a headphone terminal or a virtual device), the HRTF implementation device includes: a camera, at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the HRTF implementation method in the above-mentioned embodiment one.

[0106] Exemplarily, the HRTF implementation device can be various devices equipped with cameras, which can be Bluetooth headsets or conventional smart headsets, etc. In this case, cameras can be set in the two headsets to capture images of the wearing process of the headsets. At the same time, in order to ensure parameters such as the clarity of the captured image, depth cameras can be set in the two headsets respectively to realize image capture.

[0107] Reference below Figure 6 , which shows a schematic diagram of the structure of an implementation device of HRTF suitable for implementing the embodiment of the present application. The implementation device of HRTF in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The HRTF implementation device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0108] like Figure 6 As shown, the HRTF implementation device may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the HRTF implementation device are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following devices can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the HRTF implementation device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an HRTF implementation device with various devices, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have instead.

[0109] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0110] The HRTF implementation device provided by the present application adopts the HRTF implementation method in the above-mentioned embodiment, which can solve the technical problem of low accuracy of HRTF implementation. Compared with the prior art, the beneficial effects of the HRTF implementation device provided by the present application are the same as the beneficial effects of the HRTF implementation method provided by the above-mentioned embodiment, and other technical features in the HRTF implementation device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0111] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0112] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0113] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the HRTF implementation method in the above-mentioned embodiment.

[0114] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor devices, apparatuses or components, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution device, apparatus or component. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0115] The computer-readable storage medium may be included in the HRTF implementation device; or may exist independently without being assembled into the HRTF implementation device.

[0116] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the HRTF implementation device, the HRTF implementation device:

[0117] Acquire a human body region image captured by the headset terminal, wherein the human body region image includes a first human body region image for training to determine the transmission function and a second human body region image for matching to determine the transmission function;

[0118] Performing training according to the first human body region image to obtain first HRTF information;

[0119] Determining second HRTF information according to the second human body region image and a preset transfer function library;

[0120] A target HRTF parameter is determined according to the first HRTF information and the second HRTF information to implement the HRTF based on the target HRTF parameter.

[0121] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0122] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the device, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based device that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0123] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0124] The computer-readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned HRTF implementation method, and can solve the technical problem of low accuracy of HRTF implementation. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the HRTF implementation method provided in the above-mentioned embodiment, and will not be repeated here.

[0125] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned HRTF implementation method when executed by a processor.

[0126] The computer program product provided by this application can solve the technical problem of low accuracy of HRTF implementation. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the HRTF implementation method provided by the above embodiment, which will not be repeated here.

[0127] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for implementing HRTF, characterized in that: The HRTF implementation method is applied to a headphone terminal, and the HRTF implementation method includes: Acquire a human body region image collected by the headset terminal, wherein the human body region image includes a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function; Performing training according to the first human body region image to obtain first HRTF information; Determining second HRTF information according to the second human body region image and a preset transfer function library; A target HRTF parameter is determined according to the first HRTF information and the second HRTF information to implement the HRTF based on the target HRTF parameter.

2. The HRTF implementation method according to claim 1, characterized in that: The first human body region image includes a head image, an auricle image, and an ear canal image. The step of obtaining first HRTF information through training according to the first human body region image includes: Determine head physiological parameters in the head image, and obtain first HRTF parameters by training in a preset head training model based on the head physiological parameters; Determine the auricle physiological parameters in the auricle image, and obtain second HRTF parameters by training in a preset auricle training model based on the auricle physiological parameters; Determine the ear canal physiological parameters in the ear canal image, and obtain third HRTF parameters by training in a preset ear canal training model based on the ear canal physiological parameters, and use the first HRTF parameters, the second HRTF parameters and the third HRTF parameters as first HRTF information.

3. The method for implementing HRTF according to claim 2, characterized in that: After the step of obtaining the first HRTF information through training according to the first human body region image, the method further comprises: The first HRTF parameter, the second HRTF parameter and the third HRTF parameter are updated to a preset transfer function library.

4. The method for implementing HRTF according to claim 1, characterized in that: The second human body region image includes a torso image, and the step of determining the second HRTF information according to the second human body region image and a preset transfer function library includes: Determining a fourth HRTF parameter that matches the torso image in a preset transfer function library, and detecting whether a first matching value of the fourth HRTF parameter is greater than a preset first threshold, wherein the first matching value includes a matching degree between the torso image and the fourth HRTF parameter; When the first matching value of the fourth HRTF parameter is greater than a preset first threshold, the fourth HRTF parameter is used as the second HRTF information.

5. The method for implementing HRTF as claimed in claim 4, characterized in that: Before the step of obtaining the first HRTF information through training according to the first human body region image, the method includes: Based on the head image, the auricle image and the ear canal image in the first human body region image, the matching target HRTF parameters are determined in a preset transfer function library in turn, and a second matching value of the target HRTF parameters is detected to be greater than a preset second threshold, wherein the second matching value includes a matching degree between the head image, the auricle image or the ear canal image and the target HRTF parameters, and the second matching value is greater than the first matching value; When the second matching value of the target HRTF parameter is greater than a preset second threshold, using the target HRTF parameter as the second HRTF information; When the second matching value of the target HRTF parameter is less than or equal to a preset second threshold, the step of obtaining the first HRTF information through training according to the first human body region image is performed.

6. The method for implementing HRTF according to claim 1, characterized in that: The first HRTF information includes a first HRTF parameter of a head image, a second HRTF parameter of an auricle image, and a third HRTF parameter of an ear canal image, the second HRTF information includes a fourth HRTF parameter of a torso image, and the step of determining a target HRTF parameter according to the first HRTF information and the second HRTF information includes: A convolution sum among the first HRTF parameter, the second HRTF parameter, the third HRTF parameter and the fourth HRTF parameter is determined, and the convolution sum is used as a target HRTF parameter.

7. The method for implementing HRTF according to any one of claims 1 to 6, characterized in that: The headset terminal includes a camera and a position sensor, and the step of acquiring the human body area image collected by the headset terminal includes: Acquire the wearing process image captured by the camera and the position state captured by the position sensor in a preset first state, wherein the first state includes that the connection state of the headset terminal is the first connection and / or the input control instruction matches the preset acquisition instruction; The wearing process image is divided based on the position state and preset training features to obtain a human body region image.

8. The method for implementing HRTF according to any one of claims 1 to 6, characterized in that: The HRTF implementation method further includes: Acquiring input physiological parameters, wherein the physiological parameters include trunk physiological parameters, head physiological parameters, auricle physiological parameters and ear canal physiological parameters; Determine corresponding torso HRTF parameters based on the torso constructed physiological parameters, and determine corresponding head HRTF parameters based on the head constructed physiological parameters; Based on the physiological parameters of the auricle, the corresponding auricle HRTF parameters are determined; based on the physiological parameters of the ear canal, the corresponding ear canal HRTF parameters are determined; and the torso HRTF parameters, the head HRTF parameters, the auricle HRTF parameters and the ear canal HRTF parameters are summarized as a preset transfer function library.

9. A device for implementing HRTF, characterized in that: The HRTF implementation device is applied to a headphone terminal, and the HRTF implementation device includes: An image acquisition module, used to acquire a human body region image collected by the headset terminal, wherein the human body region image includes a first human body region image for training to determine the transfer function and a second human body region image for matching to determine the transfer function; A first processing module, configured to obtain first HRTF information by training according to the first human body region image; a second processing module, configured to determine second HRTF information according to the second human body region image and a preset transfer function library; A function determination module is used to determine target HRTF parameters according to the first HRTF information and the second HRTF information, so as to implement HRTF based on the target HRTF parameters.

10. A device for implementing HRTF, characterized in that: The HRTF implementation device includes a camera, a processor, a memory, and an HRTF implementation method program stored in the memory and executable by the processor, wherein when the HRTF implementation method program is executed by the processor, the steps of the HRTF implementation method as described in any one of claims 1 to 8 are implemented.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program for implementing a method for HRTF, wherein when the program for implementing a method for HRTF is executed by a processor, the steps of the method for implementing HRTF as claimed in any one of claims 1 to 8 are implemented.

Images