Virtual reality video adjustment system and method, storage medium and electronic equipment
By using a machine learning processing module to monitor users' physiological parameters and body movements in real time and adjusting the VR video playback scheme, the problem of insufficient immersion in existing technologies is solved, providing personalized VR video control and improving user experience and safety.
Patent Information
- Application Number
- CN202410950445.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-26
AI Technical Summary
Existing VR video control methods have low real-time adaptability to playback content, speed, viewpoint, and audio effects, which affects the user's immersion.
The system employs a machine learning processing module to monitor users' physiological parameters and limb movements in real time, and adjusts the playback scheme of virtual reality videos through machine learning algorithms, providing personalized control for low immersion, high immersion, and medical modes.
It enables personalized and interactive playback of virtual reality videos, enhancing user immersion and personalized experience, and avoiding physical injury caused by improper operation.
Smart Images

Figure CN119031192A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of virtual reality technology, and in particular to a virtual reality video adjustment system, method, storage medium and electronic device. Background Art
[0002] With the introduction of the concept of the metaverse, virtual reality (VR) technology has attracted more and more attention. It combines 360-degree panoramic video, stereo effects and a variety of interactive elements, can provide a simulated virtual environment, and enable users to obtain a highly immersive audio-visual experience. It is widely used in business, education, medical care, entertainment and other fields.
[0003] VR video viewers usually need to use physical devices such as handles or keyboards to control video playback content, speed, volume, brightness and other parameters. Since the viewer's eyes are covered by the VR device, these operations may cause inconvenience. In order to improve the convenience of VR video control, Apple has launched a gesture control solution. This method no longer requires an additional controller to adjust the playback effect of VR videos, but allows users to complete these operations through gestures. Although gesture control reduces dependence on physical controllers, viewers still need to control VR videos by themselves, which will affect the user's sense of immersion.
[0004] The existing patent with publication number CN106371605 discloses a virtual reality (VR) scene adjustment method and device, including: pre-storing VR theme scenes and determining their corresponding feature information; obtaining the film and television works selected by the user to play; determining the feature information of the selected film and television works; and determining the VR theme scene corresponding to the selected film and television works based on the feature information of the selected film and television works and the feature information of the VR theme scene. The technical solution provided by this application can increase the user's sense of immersion, and reduce the user's operation steps for selecting theme scenes before VR viewing, greatly reducing the consumption of computer performance. However, this method is limited to the change and adjustment of the playback scene environment, which makes the content, speed, viewing angle and audio effects of the playback video less adaptable to the user's real-time response during the viewing process, thereby reducing the user's sense of immersion. Summary of the invention
[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a virtual reality video adjustment system and control method.
[0006] In order to achieve the above object, the present invention adopts the following technical solution: a virtual reality video adjustment system, comprising:
[0007] A virtual reality video playback module is used to select a virtual reality video for playback;
[0008] A data storage module, used for storing virtual reality videos and their characteristic information;
[0009] The physiological monitoring module is used to monitor the physiological parameters and body movements of users in real time while watching virtual reality videos, and adjust the playback plan of virtual reality videos;
[0010] The machine learning processing module is used to extract features from the virtual reality video and push the adjusted playback parameters in real time according to the monitoring data transmitted by the physiological monitoring module;
[0011] The virtual reality video dynamic control module is used to make real-time adjustments to the currently playing virtual reality video through the instructions transmitted by the machine learning processing module.
[0012] Also included is a virtual reality video adjustment method, the specific steps of which include:
[0013] S1, the machine learning processing module uses a machine learning algorithm to analyze virtual reality videos of different themes stored in the data storage module, extracts and classifies information that causes changes in the user's physiological indicators and body movements, and constructs a relationship model;
[0014] S2, the virtual reality video playback module plays virtual reality videos of different themes according to the user's selection, and uses the physiological monitoring module to read the user's physiological indicators and body movement information to evaluate the user's activity;
[0015] S3, transmitting the information obtained by the physiological monitoring module to the machine learning processing module, and using the machine learning algorithm to compare and match the key scenes of the video and the user's activity level in real time;
[0016] S4, the machine learning processing module sends the adjustment plan to the virtual reality video dynamic control module based on the comparison information, and controls the virtual reality video playback module to output the personalized adjusted playback plan.
[0017] As a further description of the above technical solution, in step S1, it also includes:
[0018] S101: The machine learning processing module selects a virtual reality video data set as a training sample and reads the video and audio data contained in the data storage module;
[0019] S102: The machine learning processing module extracts video and audio feature information for analysis and screening;
[0020] S103: The machine learning processing module converts the feature information into a data representation suitable for processing by the machine learning algorithm;
[0021] S104: The machine learning processing module selects a machine learning algorithm to train the machine learning model according to task requirements, and constructs a relationship model between virtual reality video features and labels;
[0022] S105: The virtual reality video dynamic control module performs parameter verification and optimization on the obtained data model.
[0023] As a further description of the above technical solution, in step S2, it also includes:
[0024] S201, a physiological monitoring module collects physiological parameters of a user when watching a virtual reality video and tracks the body movements of the user during the viewing process;
[0025] S202, the machine learning processing module evaluates the user's activity level according to the physiological parameters and the amplitude of changes in body movements.
[0026] As a further description of the above technical solution, in step S3, it also includes:
[0027] S301, the machine learning processing module integrates the extracted physiological parameters, body movements and activity with the virtual reality video feature information into a feature data set;
[0028] S302, the machine learning processing module adds a corresponding label to each sample, uses the deep learning model and the integrated feature data set for training, and constructs a relationship model between each feature;
[0029] S303, the machine learning processing module evaluates the performance index of the model on the test data set;
[0030] S304, the machine learning processing module applies the trained model to the new data set, matches the user's features, and continuously optimizes the model based on feedback to improve matching accuracy.
[0031] As a further description of the above technical solution, in step S4, it also includes:
[0032] S401, a user wears a virtual reality video playback device and selects a virtual reality video to be played;
[0033] S402, the machine learning processing module switches the playback mode and sets the playback plan according to the tags contained in the virtual reality video selected by the user;
[0034] S403: the virtual reality video playback module reads the virtual reality video containing different tags from the data storage module for playback, and the physiological monitoring module monitors the user's physiological parameters in real time and tracks the user's body movements during viewing;
[0035] S404: The machine learning processing module analyzes the user's physiological parameters and body movements, identifies the user's emotional state and activity level, and compares them with the data pre-stored in the data storage module to generate a corresponding playback plan;
[0036] S405: The virtual reality video dynamic control module controls the virtual reality playback module to adjust playback parameters according to the instruction output playback scheme;
[0037] S406: Repeat S404-S405, the machine learning processing module generates a new playback plan in real time, and when the data obtained by the physiological detection module matches the pre-stored data in the machine learning processing module, stop issuing adjustment instructions to the virtual reality video dynamic control module.
[0038] As a further description of the above technical solution, the playback mode includes a low immersion mode, a high immersion mode and a medical mode.
[0039] As a further description of the above technical solution, the playback solution of the low immersion mode is:
[0040] Improve video playback quality and viewing angle, reduce video playback speed and volume, and skip or blur scenes that cause user discomfort;
[0041] The playback scheme of the high immersion mode is:
[0042] Reduce the video playback effect and field of view, increase the video playback speed and volume, and load and enhance scenes that arouse user emotions;
[0043] The playback scheme of the medical mode is:
[0044] The machine learning processing module compares the acquired physiological parameters with the pre-stored data;
[0045] When the deviation between the two is within the allowable range, no response will be made;
[0046] When the user's physiological parameters are lower than the expected data but still higher than the critical value, the machine learning processing module sends a real-time instruction to the virtual reality video dynamic control module to switch to a high immersion mode;
[0047] When the user's physiological parameters are higher than the expected data but still lower than the critical value, the machine learning processing module sends a real-time instruction to the virtual reality video dynamic control module to switch to the low immersion mode;
[0048] When the fluctuation amplitude of the user's physiological parameters is different from the pre-stored value, the machine learning processing module sends real-time instructions to the virtual reality video dynamic control module to terminate the video playback, remind the user to check his body and send a distress signal.
[0049] It also includes a computer-readable storage medium storing a computer program for the adjustment method, wherein the computer program enables a computer to execute the adjustment method as described in any one of the above technical solutions.
[0050] 10. An electronic device, comprising:
[0051] one or more processors; memory, and
[0052] One or more programs, wherein the one or more programs are stored in the memory, and the programs include a method for executing the adjustment method described in any one of the above technical solutions.
[0053] The above technical solution has the following advantages or beneficial effects:
[0054] 1. The present invention extracts and classifies feature information in virtual reality videos that may cause user reactions and changes in activity levels in advance through a machine learning algorithm, reads the user's physiological parameters and body movements in real time during video playback to evaluate the user's activity level during viewing, compares the pre-stored video feature information through a machine learning algorithm, outputs an updated playback plan, and adjusts the playback content and effect of the virtual reality video in real time, providing a personalized, interactive virtual reality video playback control solution that can provide a more immersive viewing experience in the fields of education, entertainment, and health management, and meet the personalized needs and preferences of users;
[0055] 2. According to the present invention, the user can switch between three playback modes: low immersion mode, high immersion mode and medical mode. The video playback parameters and content can be adjusted in real time by using the changes in the user's physiological parameters. This can effectively enhance the user's sense of immersion and personalized experience when using the device for watching movies or sports management, and can also effectively prevent the user from suffering physical injuries due to improper use. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 It is a schematic diagram of the virtual reality video adjustment system of the present invention;
[0057] Figure 2 The process of the virtual reality video adjustment system of the present invention Figure 1 ;
[0058] Figure 3 The process of the virtual reality video adjustment system of the present invention Figure 2 ;
[0059] Figure 4 The process of the virtual reality video adjustment system of the present invention Figure 3 ;
[0060] Figure 5The process of the virtual reality video adjustment system of the present invention Figure 4 ;
[0061] Figure 6 The process of the virtual reality video adjustment system of the present invention Figure 5 .
[0062] Legend:
[0063] 1. Virtual reality video playback module; 2. Data storage module; 3. Physiological monitoring module; 4. Machine learning processing module; 5. Virtual reality video dynamic control module. DETAILED DESCRIPTION
[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0065] Please refer to the following examples for details:
[0066] Reference Figure 1-Figure 6 The present invention provides an embodiment of a virtual reality video adjustment system, the specific structure of which includes:
[0067] Virtual reality video playback module 1, used to select a virtual reality video for playback;
[0068] Data storage module 2, used to store virtual reality video and its characteristic information;
[0069] Physiological monitoring module 3, used to monitor the user's physiological parameters (heart rate, breathing rate, blood oxygen concentration, etc.) and body movements (leaning forward, leaning back, shaking, etc.) in real time during the user's viewing of the virtual reality video, and adjust the playback plan of the virtual reality video;
[0070] The machine learning processing module 4 is used to extract features from the virtual reality video and push the adjusted playback parameters (video content, speed, viewing angle, audio effect, etc.) in real time according to the monitoring data transmitted by the physiological monitoring module 3;
[0071] The virtual reality video dynamic control module 5 is used to make real-time adjustments to the currently playing virtual reality video through the instructions transmitted by the machine learning processing module 4.
[0072] VR videos usually contain a large amount of data, including images, audio and other multi-dimensional information. Relying on manpower to identify and classify is not only inefficient but also prone to errors. Machine learning algorithms can automatically learn and extract relevant features in video data, thereby effectively classifying and identifying key data in the video and simplifying the management process of pre-stored virtual reality video information. There is no need for human intervention or manual design of feature extraction methods during the sorting process, which improves the accuracy and efficiency of video processing.
[0073] Please continue reading Figure 2-Figure 6 The present invention also provides an embodiment of a virtual reality video adjustment method, the specific steps of which include:
[0074] S1, the machine learning processing module 4 uses a machine learning algorithm to analyze virtual reality videos of different themes stored in the data storage module 2, extracts and classifies information that causes changes in the user's physiological indicators and body movements, and constructs a relationship model;
[0075] S2, the virtual reality video playback module 1 plays virtual reality videos of different themes according to the user's selection, and uses the physiological monitoring module 3 to read the user's physiological indicators and body movement information to evaluate the user's activity;
[0076] S3, transmitting the information obtained by the physiological monitoring module 3 to the machine learning processing module 4, and using the machine learning algorithm to compare and match the key scenes of the video and the user's activity level in real time;
[0077] S4, the machine learning processing module 4 sends the adjustment plan to the virtual reality video dynamic control module 5 according to the comparison information, and controls the virtual reality video playback module 1 to output the personalized adjusted playback plan.
[0078] Please continue reading Figure 2-Figure 6 , in step S1, further comprising:
[0079] S101: The machine learning processing module 4 selects a virtual reality video data set (including video files, metadata, labels and other information) as a training sample, and reads the video and audio data contained in the data storage module 2;
[0080] S102: The machine learning processing module 4 uses the extracted video and audio feature information for analysis and screening:
[0081] Among them, Convolutional Neural Networks (CNN) can be used to extract features from video frames, and optical flow method can be used to analyze motion information in video sequences. Audio features can be realized by extracting Mel-Frequency Cepstral Coefficients (MFCCs).
[0082] S103: The machine learning processing module 4 converts the feature information into a data representation suitable for processing by the machine learning algorithm:
[0083] Among them, video and audio features are converted into data representation forms such as vectors or matrices that are suitable for processing by machine learning algorithms;
[0084] S104: The machine learning processing module 4 selects a machine learning algorithm to train the machine learning model according to the task requirements, and constructs a relationship model between the virtual reality video features and the labels:
[0085] Among them, appropriate machine learning algorithms can be selected according to task requirements, such as support vector machine (SVM), random forest (RF) or deep learning (DL) model for training;
[0086] S105: The virtual reality video dynamic control module 5 performs parameter verification and optimization on the obtained data model.
[0087] In addition to extracting feature information from VR videos, machine learning algorithms can also quickly match virtual reality video features with user physiological data to design subsequent dynamic playback solutions. Figure 2-Figure 6 , in step S2, further comprising:
[0088] S201, the physiological monitoring module 3 collects the physiological parameters of the user when watching the virtual reality video and tracks the body movements of the user during the viewing process;
[0089] S202, the machine learning processing module 4 evaluates the user's activity level according to the physiological parameters and the amplitude of changes in body movements.
[0090] Please continue reading Figure 2-Figure 6 , in step S3, further comprising:
[0091] S301, the machine learning processing module 4 integrates the extracted physiological parameters, body movements and activity with the virtual reality video feature information into a feature data set;
[0092] S302, the machine learning processing module 4 adds a corresponding label to each sample, uses the deep learning model and the integrated feature data set for training, and constructs a relationship model between each feature;
[0093] S303, the machine learning processing module 4 evaluates the performance indicators of the model on the test data set:
[0094] Among them, specific indicators include accuracy, recall, precision, etc. to ensure that the model has good generalization ability;
[0095] S304, the machine learning processing module 4 applies the trained model to the new data set, matches the user's characteristics (heart rate, breathing rate, blood oxygen concentration, activity level, etc.), and continuously optimizes the model based on feedback to improve matching accuracy.
[0096] Please continue reading Figure 2-Figure 6 , in step S4, further comprising:
[0097] S401, a user wears a virtual reality video playback device and selects a virtual reality video to be played;
[0098] S402, the machine learning processing module 4 switches the play mode and sets the play plan according to the tags contained in the virtual reality video selected by the user;
[0099] S403: the virtual reality video playback module 1 reads the virtual reality video containing different tags from the data storage module 2 for playback, and the physiological monitoring module 3 monitors the user's physiological parameters in real time and tracks the user's body movements during viewing;
[0100] S404: The machine learning processing module 4 analyzes the user's physiological parameters and body movements, identifies the user's emotional state and activity level, and compares them with the data pre-stored in the data storage module 2 to generate a corresponding playback plan;
[0101] S405: The virtual reality video dynamic control module 5 controls the virtual reality playback module to adjust playback parameters according to the instruction output playback scheme;
[0102] S406: Repeat S404-S405, the machine learning processing module 4 generates a new playback plan in real time, and when the data obtained by the physiological detection module matches the pre-stored data of the machine learning processing module 4, stop issuing adjustment instructions to the virtual reality video dynamic control module 5.
[0103] Please continue reading Figure 2-Figure 6 , the playback modes include low immersion mode, high immersion mode and medical mode.
[0104] (1) The playback scheme for low immersion mode is:
[0105] Improve video playback quality and viewing angle, reduce video playback speed and volume, and skip or blur scenes that cause user discomfort;
[0106] (2) The playback scheme for high immersion mode is:
[0107] Reduce the video playback effect and field of view, increase the video playback speed and volume, and load and enhance scenes that arouse user emotions;
[0108] (3) The playback plan for the medical mode is:
[0109] The machine learning processing module 4 compares the obtained physiological parameters with the pre-stored data;
[0110] When the deviation between the two is within the allowable range, no response will be made;
[0111] When the user's physiological parameters are lower than the expected data but still higher than the critical value, the machine learning processing module 4 sends a real-time instruction to the virtual reality video dynamic control module 5 to switch to the high immersion mode;
[0112] When the user's physiological parameters are higher than the expected data but still lower than the critical value, the machine learning processing module 4 sends a real-time instruction to the virtual reality video dynamic control module 5 to switch to the low immersion mode;
[0113] When the fluctuation amplitude of the user's physiological parameters is different from the pre-stored value, the machine learning processing module 4 sends a real-time instruction to the virtual reality video dynamic control module 5 to terminate the video playback, remind the user to check his body and send a distress signal.
[0114] The present invention also provides an embodiment of a computer-readable storage medium, which stores a computer program for the adjustment method, wherein the computer program enables a computer to execute the adjustment method as described in any one of the above technical solutions.
[0115] The present invention also provides an embodiment of an electronic device, including:
[0116] one or more processors; memory, and
[0117] One or more programs, wherein the one or more programs are stored in a memory, and the programs include an adjustment method for executing any one of the above technical solutions.
[0118] Working principle:
[0119] (1) Low immersion mode, such as Figure 4 As shown:
[0120] The user wears a virtual reality video playback device and a control device with a built-in physiological parameter monitoring module;
[0121] The user chooses to play a VR video labeled "Thriller" and selects the "Low Immersion Mode" option in the effect preferences;
[0122] The virtual reality video playback module 1 reads the video data from the data storage module 2 and plays it, and at the same time starts the physiological monitoring module 3 to monitor the user's heart rate, breathing rate, blood oxygen concentration and other physiological parameters and the user's movements during the viewing process in real time;
[0123] The machine learning processing module 4 analyzes the user's physiological data and action feedback, identifies the user's emotional state and activity level, and compares them with the data pre-stored in the data storage module 2:
[0124] When it is found that the user's heart rate is higher than the expected data, the breathing frequency deviates too much from the preset data, or the blood oxygen concentration is different from the preset data, and the frequency and amplitude of the movement are also higher than the preset value, the instruction is pushed to the virtual reality video dynamic control module 5; the virtual reality video dynamic control module 5 improves the brightness, contrast and color of the virtual reality video and other playback picture effects and the user's field of view according to the instruction, reduces the video playback speed and volume, and skips or blurs the scene that causes discomfort to the user;
[0125] The physiological monitoring module 3 continues to read the user's physiological parameters after the virtual reality video is dynamically adjusted and transmits them to the machine learning processing module 4. The machine learning processing module 4 compares the obtained data with the pre-stored data again. When the two sets of data match, it stops issuing adjustment instructions to the virtual reality video dynamic control module 5.
[0126] Through the above technical solution, the user's physiological parameters and movement behaviors change during the viewing process to adjust the video playback effect, avoid the scene causing discomfort to the user, and thus improve the user experience when watching horror movies.
[0127] (2) High immersion mode, such as Figure 5 As shown:
[0128] The user wears a virtual reality video playback device and a control device with a built-in physiological parameter monitoring module.
[0129] The user chooses to play a VR video labeled "Action" and selects the "High Immersion Mode" option in the effect preferences.
[0130] The virtual reality video playback module 1 reads the video data from the storage device and plays it, and at the same time starts the physiological monitoring module 3 to monitor the user's heart rate, breathing rate, blood oxygen concentration and other physiological parameters and the user's movements during the viewing process in real time;
[0131] The machine learning processing module 4 analyzes the user's physiological data and action feedback, identifies the user's emotional state and activity level, and compares them with the data pre-stored in the storage module:
[0132] When it is found that the user's heart rate is lower than the expected data, the breathing frequency deviates greatly from the pre-stored value, or the blood oxygen concentration is different from the preset data, and the movement frequency and amplitude are also lower than the preset values, the instruction is pushed to the virtual reality video dynamic control module 5;
[0133] The virtual reality video dynamic control module 5 reduces the brightness, contrast and color of the virtual reality video and the user's field of view according to the instructions, increases the video playback speed and volume, and loads or enhances the scene that arouses the user's emotions;
[0134] The physiological monitoring module 3 continues to read the user's physiological parameters after the virtual reality video is dynamically adjusted and transmits them to the machine learning processing module 4. The machine learning processing module 4 compares the obtained data with the pre-stored data again. When the two sets of data match, it stops issuing adjustment instructions to the virtual reality video dynamic control module 5.
[0135] (3) Medical model, such as Figure 6 As shown:
[0136] The user wears the virtual reality video playback device and starts exercise training according to the scene prompts provided by the virtual reality video playback module 1;
[0137] The physiological monitoring module 3 detects the user's physiological parameters such as breathing, heart rate, blood oxygen and blood sugar in real time, and sends them to the machine learning processing module 4;
[0138] The machine learning processing module 4 compares the obtained physiological parameters with the pre-stored data:
[0139] When the deviation between the two is within the allowable range, no response will be made;
[0140] When the user's physiological parameters are lower than the expected data but still higher than the critical value, the machine learning processing module 4 sends a real-time instruction to the virtual reality video dynamic control module 5 to switch to a high immersion mode, speed up the video playback, increase the volume, or play content that is more stimulating than the current video for the user, thereby guiding the user to increase the intensity of exercise. When the expected data is reached, the current playback plan is maintained until the user finishes the exercise;
[0141] When the user's physiological parameters are higher than the expected data but still lower than the critical value, the machine learning processing module 4 sends a real-time instruction to the virtual reality video dynamic control module 5 to switch to a low immersion mode, slow down the speed of the video playback, lower the volume, or replace the video that brings strong stimulation to the user with a more peaceful content, thereby guiding the user to reduce the intensity of exercise. When the expected data is reached, the current playback plan is maintained until the user finishes the exercise;
[0142] When the amplitude of the user's physiological parameter fluctuation (increase / decrease) is different from the pre-stored value, the machine learning processing module 4 sends a real-time instruction to the virtual reality video dynamic control module 5 to terminate the video playback, remind the user to check his body and send a distress signal to avoid adverse symptoms.
[0143] Through the above technical solution, the user of this application can switch between three playback modes: low immersion mode, high immersion mode and medical mode, and use the changes in the user's physiological parameters to make real-time adjustments to the video playback parameters and content, thereby providing a personalized, interactive virtual reality video playback control solution that can effectively enhance the user's sense of immersion and personalized experience when using the device for watching movies or sports management, while also effectively preventing the user from suffering physical injuries due to improper use.
[0144] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A virtual reality video adjustment system, characterized in that: include: A virtual reality video playback module (1), used to select a virtual reality video for playback; A data storage module (2), used for storing virtual reality video and its characteristic information; A physiological monitoring module (3) is used to monitor the physiological parameters and body movements of the user in real time while watching the virtual reality video, and adjust the playback plan of the virtual reality video; A machine learning processing module (4) is used to extract features from the virtual reality video and to push the adjusted playback parameters in real time according to the monitoring data transmitted by the physiological monitoring module (3); The virtual reality video dynamic control module (5) is used to adjust the currently played virtual reality video in real time according to the instructions transmitted by the machine learning processing module (4).
2. A virtual reality video adjustment method, characterized in that: The specific steps include: S1, the machine learning processing module (4) uses a machine learning algorithm to analyze virtual reality videos of different themes stored in the data storage module (2), extracts and classifies information that causes changes in the user's physiological indicators and body movements, and constructs a relationship model; S2, the virtual reality video playback module (1) plays virtual reality videos of different themes according to the user's selection, and uses the physiological monitoring module (3) to read the user's physiological indicators and body movement information to evaluate the user's activity level; S3, transmitting the information obtained by the physiological monitoring module (3) to the machine learning processing module (4), and using the machine learning algorithm to compare and match the key scenes of the video and the user's activity level in real time; S4, the machine learning processing module (4) sends the adjustment plan to the virtual reality video dynamic control module (5) based on the comparison information, and controls the virtual reality video playback module (1) to output the personalized adjusted playback plan.
3. A virtual reality video adjustment method according to claim 2, characterized in that: In the step S1, it also includes: S101: The machine learning processing module (4) selects a virtual reality video data set as a training sample and reads the video and audio data contained in the data storage module (2); S102: The machine learning processing module (4) extracts video and audio feature information for analysis and screening; S103: The machine learning processing module (4) converts the feature information into a data representation suitable for processing by a machine learning algorithm; S104: The machine learning processing module (4) selects a machine learning algorithm according to task requirements to train the machine learning model and constructs a relationship model between virtual reality video features and labels; S105: The virtual reality video dynamic control module (5) performs parameter verification and optimization on the obtained data model.
4. A virtual reality video adjustment method according to claim 2, characterized in that: In the step S2, it also includes: S201, the physiological monitoring module (3) collects the physiological parameters of the user when watching the virtual reality video and tracks the body movements of the user during the viewing process; S202, the machine learning processing module (4) evaluates the user's activity level according to the physiological parameters and the amplitude of changes in body movements.
5. A virtual reality video adjustment method according to claim 2, characterized in that: In the step S3, it also includes: S301, the machine learning processing module (4) integrates the extracted physiological parameters, body movements and activity with the virtual reality video feature information into a feature data set; S302, the machine learning processing module (4) adds a corresponding label to each sample, uses the deep learning model and the integrated feature data set for training, and constructs a relationship model between each feature; S303, the machine learning processing module (4) evaluates the performance index of the model on the test data set; S304, the machine learning processing module (4) applies the trained model to the new data set, matches the user's features, and continuously optimizes the model based on feedback to improve matching accuracy.
6. A virtual reality video adjustment method according to claim 2, characterized in that: In the step S4, it also includes: S401, a user wears a virtual reality video playback device and selects a virtual reality video to be played; S402, the machine learning processing module (4) switches the playback mode and sets the playback plan according to the tags contained in the virtual reality video selected by the user; S403: the virtual reality video playback module (1) reads the virtual reality video containing different tags from the data storage module (2) for playback, and the physiological monitoring module (3) monitors the user's physiological parameters in real time and tracks the user's body movements during the viewing process; S404: The machine learning processing module (4) analyzes the user's physiological parameters and body movements, identifies the user's emotional state and activity level, and compares them with the data pre-stored in the data storage module (2) to generate a corresponding playback plan; S405: The virtual reality video dynamic control module (5) controls the virtual reality playback module to adjust playback parameters according to the instruction output playback scheme; S406: Repeat S404-S405, the machine learning processing module (4) generates a new playback plan in real time, and when the data obtained by the physiological detection module matches the pre-stored data of the machine learning processing module (4), the adjustment instruction to the virtual reality video dynamic control module (5) is stopped.
7. A virtual reality video adjustment method according to claim 6, characterized in that: The playback modes include a low immersion mode, a high immersion mode and a medical mode.
8. A virtual reality video adjustment method according to claim 7, characterized in that: The playback scheme of the low immersion mode is: Improve video playback quality and viewing angle, reduce video playback speed and volume, and skip or blur scenes that cause user discomfort; The playback scheme of the high immersion mode is: Reduce the video playback effect and field of view, increase the video playback speed and volume, and load and enhance scenes that arouse user emotions; The playback scheme of the medical mode is: The machine learning processing module (4) compares the obtained physiological parameters with pre-stored data; When the deviation between the two is within the allowable range, no response will be made; When the user's physiological parameters are lower than expected data but still higher than a critical value, the machine learning processing module (4) sends a real-time instruction to the virtual reality video dynamic control module (5) to switch to a high immersion mode; When the user's physiological parameters are higher than expected data but still lower than the critical value, the machine learning processing module (4) sends a real-time instruction to the virtual reality video dynamic control module (5) to switch to a low immersion mode; When the fluctuation amplitude of the user's physiological parameters is different from the pre-stored value, the machine learning processing module (4) sends a real-time instruction to the virtual reality video dynamic control module (5) to terminate the video playback, remind the user to check his body and send a distress signal.
9. A computer-readable storage medium, characterized in that: It stores a computer program for the adjustment method, wherein the computer program enables a computer to execute the adjustment method according to any one of claims 2 to 8.
10. An electronic device, characterized in that: include: one or more processors; Memory, and One or more programs, wherein the one or more programs are stored in the memory, and the programs include instructions for executing the adjustment method according to any one of claims 2 to 8.
Citation Information
Patent Citations
Method of adjusting parameters of virtual-reality device and virtual-reality devices
CN107491166A
Virtual reality emotion control method and device
CN113485557A
Equipment use experience monitoring method and device, equipment and storage medium
CN116269385A
Interaction control method and apparatus for virtual reality device, and device
WO2023045031A1
Cited By
Intelligent sports event playing method based on VR
CN120223913A
Display equipment control method and device, electronic equipment and storage medium
CN121037629A
Digital human navigation type marketing short video generation method and system
CN121691842A
