A method for matching posture data in VR cloud rendering

Abstract

The present invention discloses a method for VR cloud rendering matching posture data, comprising the following steps: obtaining the initial posture data of a VR helmet in real time, and setting a posture ID for it; sending the initial posture data and its corresponding posture ID to a server, and the server submitting the initial posture data to a VR application; the VR application performs a rendering operation, and generates a video image and rendering posture data; the server obtains the generated video image and rendering posture data, and calculates the deviation value α between the initial posture data and the rendering posture data saved on the server; determines whether the deviation value α is less than a deviation threshold; if less than, the initial posture data matches the rendering posture data. The present invention can achieve rapid matching of the video image generated after rendering and the initial posture data corresponding thereto by calculating the deviation value between the rendering posture data and the initial posture data and comparing it with a preset threshold, find the best initial posture data, and ensure good local rendering effect.

Description

Technical Field

[0001] The present invention relates to the field of VR cloud rendering, and in particular to a method for matching posture data of VR cloud rendering. Background Art

[0002] VR applications run on cloud servers. VR helmets are connected to cloud servers through the network and send posture data required for cloud rendering to the server. After receiving the posture data, the server renders through the VR application. After the rendering is completed, the cloud server transmits the video data back to the local VR helmet for rendering and display. Since the local rendering and display requires the corresponding posture data, it is necessary to match the posture data to select the best posture data. In the prior art, it is impossible to quickly match the VR rendering posture data, resulting in the inability to match the best posture data, which results in the rendered image being not smooth and the rendering effect being poor.

[0003] Therefore, the prior art has defects and needs to be improved. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for VR cloud rendering matching posture data to solve the problem that the prior art cannot achieve fast matching of VR rendering posture data, resulting in uneven rendered images and ultimately poor rendering efficiency.

[0005] The technical solution of the present invention is as follows: A method for VR cloud rendering matching posture data comprises the following steps:

[0006] S1: The VR driver client obtains a number of initial posture data of the VR helmet in real time, sets a posture ID for each initial posture data, and binds and saves the initial posture data with its corresponding posture ID.

[0007] S2: The VR driver client sends the acquired initial posture data and the posture ID corresponding to the initial posture data to the server in real time. After saving the initial posture data and the corresponding posture ID, the server submits a set of initial posture data to the VR application.

[0008] S3: The VR application performs rendering operations according to the corresponding initial posture data, and generates corresponding video images and rendering posture data after the rendering is completed.

[0009] S4: The server obtains the video image and rendering posture data generated after rendering on the VR application, and calculates the deviation value α between the initial posture data saved on the server and the rendering posture data.

[0010] The deviation value α=(c11) 2 +(c12)2 +…+(c33) 2 +(c34) 2 ;

[0011]

[0012] Among them, the b 11 、b 12 …b 34 is the data value in the rendering posture data, the a 11 、a 12 …a 34 is the data value in the initial posture data; the initial posture data and the rendering posture data are both 3*4 matrices.

[0013] S5: Determine whether the deviation value α is less than a deviation threshold set on the server.

[0014] S6: If the deviation value α is less than the deviation threshold, the initial posture data matches the rendering posture data, and the posture ID of the initial posture data corresponding to the deviation value α is obtained, and the posture ID and the video image generated in step S3 are sent to the VR driver client. The VR driver client searches for the initial posture data corresponding to the posture ID according to the received posture ID, and submits the found initial posture data and the video image received by the VR driver client to the VR helmet for local rendering and display.

[0015] Furthermore, after step S6, the following steps are further included:

[0016] S7: If the deviation value α is not less than the deviation threshold, then calculate the deviation value α between the next set of initial posture data and the rendering posture data, and repeat steps S5 to S6 until initial posture data matching the rendering posture data is found.

[0017] Furthermore, in step 3, after the VR application completes rendering, the server submits the next set of initial posture data to the VR application.

[0018] Furthermore, in step S6, after obtaining the posture ID of the initial posture data corresponding to the deviation value α, the server will encode and compress the video image generated in step S3, and send the posture ID and the encoded and compressed video image together to the VR driver client.

[0019] Furthermore, the VR application is a vrcar program or a FruitNinjaVR program.

[0020] Furthermore, in step S4, the server obtains the video image and rendering posture data generated after rendering on the VR application through the openvr interface.

[0021] By adopting the above scheme, the present invention provides a method for VR cloud rendering matching posture data, which has the following beneficial effects: by calculating the deviation value α between the rendering posture data and the initial posture data, and comparing it with a preset threshold, it is possible to achieve a fast matching of the video image generated after rendering and the initial posture data corresponding to it, find an optimal initial posture data, and perform local rendering through the initial posture data and the video image rendered on the server. The rendered image is displayed very smoothly, ensuring a good local rendering effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the structure of the present invention. DETAILED DESCRIPTION

[0023] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0024] Please refer to Figure 1 The present invention provides a method for VR cloud rendering matching posture data, comprising the following steps:

[0025] S1: The VR driver client obtains several initial posture data of the VR helmet in real time, sets a posture ID for each initial posture data, and binds and saves the initial posture data with its corresponding posture ID. The VR driver client will obtain the initial posture data on the VR helmet in real time, and create a posture ID corresponding to the initial posture data, so that each initial posture data corresponds to a posture ID. After obtaining and creating the corresponding posture ID, multiple groups of initial posture data and their respective corresponding posture IDs will be sent to the server in real time. In addition, in step S1, the frequency at which the VR driver client obtains the initial posture data on the VR helmet is generally fixed to 3 times the video frame rate. The video frame rate is set when the VR driver client is initialized, for example, 72fps.

[0026] S2: The VR driver client sends the acquired initial posture data and the posture ID corresponding to the initial posture data to the server in real time. After saving the initial posture data and the corresponding posture ID, the server submits a set of initial posture data to the VR application. There are a variety of VR applications on the server, such as the vrcar program and the FruitNinjaVR program, which can be set according to actual needs; in addition, when the server submits the initial posture data to the VR application, it submits it to the corresponding VR application according to the type of different initial posture data, which is specifically regulated by the server.

[0027] S3: The VR application performs rendering operations according to the corresponding initial posture data, and generates corresponding video images and rendering posture data after the rendering is completed. After the VR application completes the rendering, the server submits the next set of initial posture data to the VR application for rendering.

[0028] S4: The server obtains the video image and rendering posture data generated after rendering on the VR application, and calculates the deviation value α between the initial posture data saved on the server and the rendering posture data. Specifically, the server obtains the video image and rendering posture data generated after rendering on the VR application through the openvr interface.

[0029] The deviation value α=(c11) 2 +(c12) 2 +…+(c33) 2 +(c34) 2 ;

[0030]

[0031] Among them, the b 11 、b 12 …b 34 is the data value in the rendering posture data, the a 11 、a 12 …a 34 is the data value in the initial posture data; the initial posture data and the rendering posture data are both 3*4 matrices.

[0032] S5: Determine whether the deviation value α is less than a deviation threshold set on the server. The deviation threshold is a value set in advance and can be set according to actual needs.

[0033] S6: If the deviation value α is less than the deviation threshold, the initial posture data matches the rendering posture data, and the posture ID of the initial posture data corresponding to the deviation value α is obtained, and the posture ID and the video image generated in step S3 are sent to the VR driver client. The VR driver client searches for the initial posture data corresponding to the posture ID according to the received posture ID, and submits the found initial posture data and the video image received by the VR driver client to the VR helmet for local rendering and display.

[0034] A method for matching posture data of VR cloud rendering in the present invention sets a posture ID corresponding to the initial posture data obtained from the VR helmet by the VR driver client, so that each initial posture data corresponds to a posture ID, and sends the obtained initial posture data and the posture ID corresponding to each initial posture data to the server. After receiving the initial posture data and the corresponding posture ID, the server saves them and submits the initial posture data to the VR application for rendering in sequence. After completing the rendering of a set of initial posture data, the server immediately sends the next set of initial posture data to the VR application. It should be noted that the VR driver client also obtains other posture data from the VR helmet at this time and repeats the above operation. At the same time, the server obtains the video image and rendering posture data generated after the VR application is rendered, and calculates the rendering posture data and the current initial posture data on the server (all initial posture data). The deviation value α of the initial posture data is calculated in turn, and the deviation value α is compared with a preset deviation threshold. If the deviation value α is less than the preset deviation threshold, it indicates that the initial posture data matches the current rendering posture data, and the posture ID corresponding to the initial posture data is found. The posture ID and the generated video image are sent to the VR driver client. After receiving the video image and the posture ID, the VR driver client can quickly find the initial posture data corresponding to it according to the posture ID, and send the video image and the corresponding initial posture data to the VR helmet for body rendering; by calculating a method for matching posture data of VR cloud rendering in the present invention, a fast matching of the video image generated after rendering and the initial posture data corresponding to it can be achieved, and an optimal initial posture data is found. Local rendering is performed through the initial posture data and the video image rendered on the server. The rendered image is displayed very smoothly, ensuring a good local rendering effect.

[0035] Specifically, in this embodiment, after step S6, the following steps are further included:

[0036] S7: If the deviation value α is not less than the deviation threshold, then calculate the deviation value α between the next set of initial posture data and the rendering posture data, and repeat steps S5 to S6 until the initial posture data matching the rendering posture data is found. Since there are multiple sets of data on the server, it may be necessary to calculate multiple sets of data to find the initial posture data matching the rendering posture data; in addition, if all the initial posture data on the server are traversed and no initial posture data that meets the requirements is matched, the value of the posture ID is set to 0, and the server sends the posture ID and the rendered video image to the VR driver client. The VR driver client checks the posture ID and finds that it is 0, and then uses the initial posture data of the last local rendering for rendering.

[0037] Specifically, in this embodiment, in step S6, after obtaining the posture ID of the initial posture data corresponding to the deviation value α, the server will encode and compress the video image generated in step S3, and send the posture ID and the encoded and compressed video image together to the VR driver client.

[0038] In summary, the present invention provides a method for VR cloud rendering matching posture data, which has the following beneficial effects: by calculating the deviation value α between the rendering posture data and the initial posture data, and comparing it with a preset threshold, it is possible to achieve a fast matching of the video image generated after rendering and the initial posture data corresponding to it, find an optimal initial posture data, and perform local rendering through the initial posture data and the video image rendered on the server. The rendered image is displayed very smoothly, ensuring a good local rendering effect.

[0039] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for matching posture data for VR cloud rendering, characterized in that: The following steps are involved: S1: The VR driver client obtains a number of initial posture data of the VR helmet in real time, sets a posture ID for each initial posture data, and binds and saves the initial posture data with its corresponding posture ID; S2: The VR driver client sends the acquired initial posture data and the posture ID corresponding to the initial posture data to the server in real time. After saving the initial posture data and the corresponding posture ID, the server submits a set of initial posture data to the VR application. S3: The VR application performs rendering operations according to the corresponding initial posture data, and generates corresponding video images and rendering posture data after the rendering is completed; S4: The server obtains the video image and rendering posture data generated after rendering on the VR application, and calculates the deviation value α between the initial posture data saved on the server and the rendering posture data; The deviation value α=(c11) 2 + (c12) 2 + … + (c33) 2 + (c34) 2 ; Among them, the b 11 、b 12 …b 34 is the data value in the rendering posture data, the a 11 、a 12 …a 34 is the data value in the initial posture data; both the initial posture data and the rendering posture data are 3*4 matrices; S5: Determine whether the deviation value α is less than a deviation threshold set on the server; S6: If the deviation value α is less than the deviation threshold, the initial posture data matches the rendering posture data, and a posture ID of the initial posture data corresponding to the deviation value α is obtained, and the posture ID and the video image generated in step S3 are sent to the VR driver client. The VR driver client searches for the initial posture data corresponding to the posture ID according to the received posture ID, and submits the searched initial posture data and the video image received by the VR driver client to the VR helmet for local rendering and display; After step S6, the following steps are also included: S7: If the deviation value α is not less than the deviation threshold, then calculating the deviation value α between the next set of initial posture data and the rendering posture data, and repeating steps S5 to S6 until initial posture data matching the rendering posture data is found; In step S6, after obtaining the posture ID of the initial posture data corresponding to the deviation value α, the server will encode and compress the video image generated in step S3, and send the posture ID and the encoded and compressed video image together to the VR driver client.

2. A method for matching posture data of VR cloud rendering according to claim 1, characterized in that: In step S3, after the VR application completes rendering, the server submits the next set of initial posture data to the VR application.

3. The method for matching posture data of VR cloud rendering according to claim 1, characterized in that: The VR application is a vrcar program or a FruitNinjaVR program.

4. The method for matching posture data of VR cloud rendering according to claim 1, characterized in that: In step S4, the server obtains the video image and rendering posture data generated after rendering on the VR application through the openvr interface.

Images