Driving test teaching content generation method and system

Abstract

The embodiment of the invention provides a driving test teaching content generation method and system, and can solve the problem that in order to carry out teaching propagation, a coach will create a teaching video, but the coach needs to have the capabilities of video production and copywriting, and needs to spend more time and energy, so that the coach is difficult to efficiently output professional video content. The method comprises the following steps: acquiring historical teaching video data and image-text content data issued by a target coach user on a driving test teaching platform; generating target teaching video data through an AI large model based on the historical teaching video data and the image-text content data, wherein the target teaching video data is associated with styles of the historical teaching video data and the image-text content data of the target coach user; and uploading the generated target teaching video data to a social platform bound with the target coach user, and publishing the target teaching video data on the social platform by a user account associated with the target coach user on the social platform.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and system for generating driving test teaching content. Background Art

[0002] On driving test training platforms, users can share their driving test experiences, ask questions, or seek help. At the same time, some driving test instructors also post instructional videos on the platform to help students better understand driving skills and test key points. However, high-quality instructional videos require not only professional driving instructional skills from instructors, but also video production and copywriting skills. The production process of traditional instructional videos involves multiple steps, including scriptwriting, recording, editing, and adding subtitles, which is time-consuming and labor-intensive. This not only increases the instructor's workload but also requires a significant investment of their time and energy, thereby reducing the likelihood of efficiently delivering professional instructional content. This situation can lead to a shortage of high-quality teaching resources, impacting students' learning experience and test preparation efficiency. Summary of the Invention

[0003] The embodiments of the present application provide a method and system for generating driving test teaching content, which can solve the problem that coaches create teaching videos for the purpose of teaching dissemination, but this requires coaches to have the ability to produce videos and write copy, and it takes a lot of time and energy, making it difficult for coaches to efficiently output professional video content.

[0004] A first aspect of an embodiment of the present application provides a method for generating driving test teaching content, comprising:

[0005] Obtain historical teaching video data and graphic content data published by the target instructor user on the driving test teaching platform;

[0006] Generate target teaching video data based on the historical teaching video data and graphic content data through the AI big model, wherein the target teaching video data is associated with the style of the historical teaching video data and graphic content data of the target coach user;

[0007] The generated target teaching video data is uploaded to the social platform bound to the target coach user and published on the social platform using the user account associated with the target coach user on the social platform.

[0008] Optionally, it also includes:

[0009] Obtaining video data uploaded by the target coach user to analyze the video data;

[0010] In the case where it is determined through analysis that the video data includes external teaching video data, extracting position parameters from metadata of the external teaching video data;

[0011] Query the 3D map database based on the position parameter to obtain the corresponding shooting driving trajectory of the off-road teaching video data in the 3D map;

[0012] Fuse the shooting driving trajectory into the picture of the off-road teaching video data to generate multi-perspective teaching video data, and upload the multi-perspective teaching video data to the driving test teaching platform after confirmation by the target coach user. The multi-perspective teaching video data includes the shooting perspective and the shooting driving trajectory perspective of the off-road teaching video data.

[0013] Optionally, it further includes:

[0014] When it is analyzed and determined that the video data includes off-road teaching video data, extract the timestamp information in the metadata of the off-road teaching video data;

[0015] Establish a time mapping between the shooting driving trajectory and the off-road teaching video data based on the timestamp information, so that the shooting driving trajectory in the multi-perspective teaching video data and the video frames in the off-road teaching video data are displayed corresponding to each other based on the time mapping.

[0016] Optionally, it further includes:

[0017] Query the real-scene map database based on the position parameter to generate real-scene driving simulation data according to the shooting driving trajectory in combination with the real-scene map database;

[0018] Fuse the real-scene driving simulation data into the picture of the off-road teaching video data to generate multi-perspective teaching video data.

[0019] Optionally, the shooting driving trajectory is displayed in the multi-perspective teaching video data in the form of a window.

[0020] The second aspect of the embodiments of the present application provides a driving test teaching content generation device, including:

[0021] An acquisition unit, configured to acquire historical teaching video data and graphic and text content data published by a target coach user on a driving test teaching platform;

[0022] A generation unit generates target teaching video data based on the historical teaching video data and the graphic and text content data through an AI large model, and the target teaching video data is associated with the styles of the historical teaching video data and the graphic and text content data of the target coach user;

[0023] Upload the generated target teaching video data to the social platform bound to the target coach user and publish it on the social platform with the user account associated with the target coach user on the social platform.

[0024] In the third aspect of the embodiments of the present application, an electronic system is provided, including a memory and a processor, and the processor is configured to implement the steps of the above-mentioned driving test teaching content generation method when executing a computer program stored in the memory.

[0025] In the fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program implements the steps of the above-mentioned driving test teaching content generation method when executed by a processor.

[0026] In summary, for the driving test teaching content generation method provided by the embodiments of the present application, historical teaching video data and graphic and text content data published by a target coach user on a driving test teaching platform are obtained; based on the historical teaching video data and graphic and text content data, target teaching video data is generated through an AI large model, and the target teaching video data is associated with the styles of the historical teaching video data and graphic and text content data of the target coach user; the generated target teaching video data is uploaded to a social platform bound to the target coach user and published on the social platform using the user account associated with the target coach user on the social platform. Thus, the AI automatically generates teaching content, eliminating the need for coaches to manually edit, dub, and compile, saving a significant amount of time and improving teaching efficiency. The AI learns the language habits and teaching logic of the coaches, making the generated content highly consistent with the coaches' styles and enhancing the brand influence. Compared with traditional manually produced videos, the AI can quickly generate and regularly publish content, increasing the content activity of the platform. Combining technologies such as 3D animations, automatic subtitles, and voice explanations makes the teaching videos more intuitive and easier to understand, improving the passing rate of students. Through automated multi-platform publishing, the teaching coverage of coaches is expanded, attracting more students' attention and enhancing market competitiveness.

[0027] Correspondingly, the driving test teaching content generation device, electronic system, and computer-readable storage medium provided by the embodiments of the present invention also have the above-mentioned technical effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic flowchart of a possible driving test teaching content generation method provided by an embodiment of the present application;

[0029] Figure 2 It is a schematic structural block diagram of a possible driving test teaching content generation device provided by an embodiment of the present application;

[0030] Figure 3 It is a schematic hardware structure diagram of a possible driving test teaching content generation device provided by an embodiment of the present application;

[0031] Figure 4 It is a schematic structural block diagram of a possible electronic system provided by an embodiment of the present application;

[0032] Figure 5 It is a schematic structural block diagram of a possible computer-readable storage medium provided by an embodiment of the present application. Detailed implementation manners

[0033] An embodiment of the present application provides a method and system for generating driving test teaching content, which can solve the problem that in order to carry out teaching dissemination, coaches will create teaching videos, but this requires coaches to have the ability of video production and copywriting, and it takes a lot of time and energy, making it difficult for coaches to efficiently output professional video content.

[0034] Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0035] Please refer to Figure 1 , which is a flowchart of a method for generating driving test teaching content provided by an embodiment of the present application, and specifically may include: S110 - S130.

[0036] S110, obtain historical teaching video data and graphic content data published by a target coach user on a driving test teaching platform.

[0037] S120, generate target teaching video data based on the historical teaching video data and graphic content data through an AI large model, and the target teaching video data is associated with the styles of the historical teaching video data and graphic content data of the target coach user.

[0038] S130, upload the generated target teaching video data to a social platform bound to the target coach user and publish it on the social platform with the user account associated with the target coach user on the social platform.

[0039] It is understandable that the core idea of this solution is to use large AI models to automatically generate high-quality driving test teaching videos, reducing the content production burden on coaches and maintaining their personalized teaching styles. By collecting the historical teaching videos, pictures and texts posted by target coach users on the driving test teaching platform, the large AI model can learn the teaching styles of coaches, including tone, expression, teaching logic, etc. Based on the learned coach styles, the AI model automatically generates new teaching videos, including pictures, explanatory voices, subtitles and animation demonstrations, making the generated content not only conform to the coach's style but also efficiently convey teaching information. The generated videos can be directly uploaded to the social platform bound to the coach and automatically posted in the name of the coach's social account, enhancing the coach's influence and student coverage.

[0040] Exemplarily, the system needs to collect the historical teaching videos, pictures and texts posted by target coach users on the driving test teaching platform and analyze this data to establish a teaching style model for the coach. First, obtain the coach's past video data through web crawling or API interfaces, including teaching explanations for Subject 2 and Subject 3, test skill analysis, operation demonstrations, etc. At the same time, extract picture and text data such as articles, teaching notes, social media posts, and user interaction comments posted by the coach on the platform to analyze their word usage habits, expression methods, and teaching logic. During the data processing, frame extraction and key frame recognition are performed on the video content to analyze the teaching style in the video, such as whether gesture demonstrations are used and whether animations are coordinated. For the audio part, features such as the coach's speaking speed, intonation, and stress are obtained through speech-to-text technology (ASR), and combined with natural language processing (NLP) technology to perform sentiment analysis on the picture and text data, summarizing the coach's explanation logic, such as whether they tend to use humorous metaphors or directly elaborate on theoretical knowledge. Subsequently, use deep learning models (such as Transformer or GPT) to train this data to build a dedicated coach style generation model, enabling the AI to have the ability to imitate the coach's style. For example, if a certain coach is used to using vivid metaphors when explaining "Subject 2 reverse parking", such as "It's like threading a needle, the rear of the car is the eye of the needle, and the front of the car is the thread", the AI model will learn this expression style and apply similar metaphors in the generated teaching content to maintain the coach's unique style.

[0041] Exemplarily, based on the trained coach style model, the large AI model will automatically generate new teaching videos, including multiple elements such as complete teaching explanations, video footage, subtitles, etc. First, the system will combine the data analysis of the platform to determine the popular questions that trainees are concerned about, such as "how to accurately control the direction when reversing into the parking space". The AI will automatically search for relevant past content of the coach and generate an optimized teaching explanation script in combination with the latest driving test rules. In the voice synthesis session, the AI will use text-to-speech (TTS) technology, such as Edge TTS or Google TTS, to generate an explanatory audio that matches the coach's voice style, ensuring that the intonation and rhythm are consistent with the original teaching style. For the video footage, the AI can be generated in various ways. For example, in 3D animation demonstrations, the AI can generate 3D simulation animations, such as the trajectory changes when reversing into the parking space, and the lighting simulation operations in the third subject road test. Or virtual narrators, digital humans (AI-synthesized coach images) can be used for teaching to enhance the interactivity of teaching. Or automatic subtitle matching, the AI will analyze the voice timeline and intelligently match the subtitles to synchronize them with the voice explanation, improving the learning experience. Suppose the AI needs to generate a teaching video for "parallel parking in subject two". The system will automatically call up 3D animations to show the whole process of the car entering the parking space from the initial position. At the same time, the generated voice explanation may be: "First, turn the steering wheel all the way to the right, observe the rearview mirror, and straighten the steering wheel when the body is parallel to the warehouse line." The subtitles will automatically match the voice and appear dynamically to improve readability.

[0042] Exemplarily, when the AI successfully generates a teaching video, the system will automatically optimize the video format and upload it to the social platform bound to the target coach user to ensure an efficient and error-free publishing process. First, the system will automatically adjust the video resolution and ratio according to the requirements of different social platforms. Next, the AI will analyze the coach's social media publishing habits. For example, if a certain coach is used to posting videos on Wednesdays and Fridays, the system will schedule the upload according to this rhythm. In addition, the AI can also optimize the title and tags to make the content more attractive. For example, the title: "100% Passing Skills for Reversing into the Parking Space in Subject Two, Master the Key Points in 3 Minutes!" Keyword tags: "#DrivingTest#ReversingIntoParkingSpace#DrivingTestSkills#Must-See for Newbies" This not only improves the recommendation weight of the video but also attracts more targeted users to watch. For example, if the coach is used to posting teaching content on Bilibili, the AI will automatically adjust the video format and use Bilibili's API interface for uploading to ensure that the content can seamlessly integrate into the social media ecosystem and be published in the name of the coach, continuously expanding the influence of their teaching.

[0043] Exemplarily, after the video is released, the system will continuously monitor the interactive feedback of the trainees, and combine the AI analysis results to optimize the subsequent teaching content. The AI will automatically collect user comments and extract high-frequency questions, such as "How should the rearview mirror be adjusted when reversing?" and "How to avoid stalling in the third subject?" For these questions, the AI will intelligently recommend relevant existing videos of the coach or generate new teaching content for answers. In addition, the AI can also analyze the viewing data of the videos, such as the playing duration, like rate, and forwarding volume, etc., to determine which type of content is more popular among the trainees and adjust the generation strategy of subsequent teaching videos. For example, if the videos in the "Skills of Road Test in the Third Subject" series have a high playback volume, the AI may generate a more detailed step-by-step explanation video, such as "Detailed Interpretation of Nighttime Lighting Simulation Test", so as to accurately meet the needs of the trainees. For example, after a coach released a video on "Skills of Hill Start", the AI detected that there were many questions like "How to prevent the vehicle from rolling back?" in the user comments. Then the system automatically generated a new video, which detailed the operation skills to prevent the vehicle from rolling back, and replied to the users in the comment area, recommending this video to enhance user stickiness.

[0044] In summary, for the method for generating driving test teaching content provided by the above embodiments, historical teaching video data and graphic and text content data released by a target coach user on a driving test teaching platform are obtained; based on the historical teaching video data and graphic and text content data, target teaching video data is generated through an AI large model, and the target teaching video data is associated with the styles of the historical teaching video data and graphic and text content data of the target coach user; the generated target teaching video data is uploaded to a social platform bound to the target coach user and is published on the social platform using the user account associated with the target coach user on the social platform. Thus, the AI automatically generates teaching content, and the coach does not need to manually edit, dub, and edit, which greatly saves time and improves teaching efficiency. The AI learns the language habits and teaching logic of the coach, making the generated content highly consistent with the coach's style and enhancing the brand influence. Compared with traditional hand-made videos, the AI can quickly generate and regularly publish, improving the content activity of the platform. Combining technologies such as 3D animation, automatic subtitles, and voice explanations makes the teaching videos more intuitive and easy to understand, improving the passing rate of the trainees. Through automated multi-platform publishing, the teaching coverage of the coach is expanded, attracting more trainee attention and improving market competitiveness.

[0045] In one embodiment, it further includes:

[0046] Obtain the video data uploaded by the target coach user to analyze the video data;

[0047] In the case where it is analyzed and determined that the video data includes out-road teaching video data, extract the position parameters in the metadata of the out-road teaching video data;

[0048] Query the 3D map database based on the position parameter to obtain the corresponding shooting driving trajectory of the off-road teaching video data in the 3D map;

[0049] Fuse the shooting driving trajectory into the picture of the off-road teaching video data to generate multi-perspective teaching video data, and upload the multi-perspective teaching video data to the driving test teaching platform after the target coach user confirms it. The multi-perspective teaching video data includes the shooting perspective and the shooting driving trajectory perspective of the off-road teaching video data.

[0050] It can be understood that based on the AI automatic generation of teaching videos, this solution further enhances the visualization effect of off-road teaching videos. Specifically, by analyzing the videos uploaded by the target coach user, identifying the off-road teaching content therein, extracting the position information (such as GPS coordinates) of the videos, and then querying the corresponding driving trajectories in combination with the 3D map database, the trajectory data is fused with the original video picture to generate multi-perspective teaching videos. By showing the driving route from multiple perspectives (including the original driving perspective + the 3D trajectory perspective), it enables students to more clearly understand the road environment and driving operations. Combining real videos and virtual 3D map perspectives allows students to understand driving operations from different angles and strengthens their memory of the test route. Using AI and automated data processing enables coaches to quickly generate professional multi-perspective teaching videos without manual trajectory annotation.

[0051] Exemplarily, when a coach user uploads a teaching video to the driving test teaching platform, the system will automatically analyze the video content to determine whether the video belongs to an off-road teaching video. First, use computer vision technologies (such as YOLO, OpenCV) to identify the scene in the video and determine whether it contains features of the off-road driving environment such as road signs, traffic lights, and vehicle driving information. If the video content contains significant road elements (such as lane lines, traffic lights, etc.), it is further confirmed that the video belongs to an off-road teaching video. In addition, the system will also combine the audio data of the video and use speech recognition (ASR) technology to analyze the coach's speech explanation content, such as whether terms related to off-road driving such as "lane change", "U-turn at the intersection", and "traffic light passing rules" are mentioned, to improve the accuracy of the judgment. For example, a certain video uploaded by a coach contains an explanation of "the skills of passing traffic lights". The system detects traffic lights and lane markings in the picture through video analysis, and at the same time, the speech recognition recognizes a statement such as "When going straight through the intersection, pay attention to observing the vehicles on the left and right". The system confirms that the video belongs to an off-road teaching video.

[0052] Exemplarily, after determining that the video belongs to an external road teaching video, the system will further extract the location information (GPS coordinates) in the metadata of the video. Usually, videos taken by smartphones or dash cams will automatically record data such as GPS coordinates and timestamps, and the system can obtain the corresponding driving trajectory of the video by reading this metadata. If the location information is missing in the metadata of the video, the system can also use computer vision technology (such as SLAM algorithm) combined with map data to infer the road location where the video was taken. For example, the video taken by the coach's dash cam contains GPS coordinates (such as longitude: 116.403874, latitude: 39.914889), and the system extracts this location information, indicating that the video was taken on a certain road in Beijing.

[0053] Exemplarily, after obtaining the location information, the system will call a 3D map database (such as xx Map), query the corresponding road structure based on the GPS coordinates, and generate the driving trajectory of the teaching video. This process usually includes: matching road information, finding the corresponding road name, lane information, traffic light positions, etc. according to the GPS coordinates; reconstructing the driving trajectory, using the map API to query the historical traffic data and road topology of the area, and accurately restoring the driving path of the vehicle; and enhancing the trajectory data, combining 3D map data to optimize the trajectory display and make it more intuitively presented to the trainees. For example, if the GPS coordinates of a certain video correspond to "Chang'an Avenue in Beijing", the system will query the 3D road data of this area and draw the complete driving trajectory of the coach, including key information such as the starting point, ending point, turning points, and traffic lights.

[0054] Exemplarily, after obtaining the driving trajectory, the system will fuse it with the original teaching video to generate a multi-view teaching video. The main fusion methods can include: Picture-in-Picture (PIP) mode, adding a small window in the lower right or left corner of the original video to display the driving trajectory in the 3D map, so that the trainees can see the driving picture and the trajectory indication at the same time. Split-screen mode, dividing the screen into two parts, one part shows the original driving perspective of the coach, and the other part shows the driving route in the 3D map, helping the trainees better understand the relationship between driving operations and the road environment. Or, the trajectory overlay mode, directly overlaying a semi-transparent driving trajectory on the driving picture of the original video and marking key sections (such as U-turn points and lane-changing points on the exam route), enhancing the intuitiveness. For example, the coach uploads a video of the third subject road test route, and the system automatically displays the driving trajectory of this route in the 3D map on the left side of the video screen, marks the traffic lights at important intersections, and indicates the driving direction of the coach with animated arrows, making it easier for the trainees to understand this exam route.

[0055] For example, after a multi-view instructional video is generated, the system automatically provides a preview interface for the target instructor to review. If the instructor is satisfied with the video content, they can upload it directly to the driving test teaching platform with a single click for students to learn from. The system also allows instructors to make fine-tuning adjustments, such as adjusting the transparency of the track, modifying annotations, and adding additional audio commentary, to ensure optimal teaching results. For example, after the instructor confirms the video and clicks "Confirm Publish," the system uploads the multi-view instructional video to the "External Road Test Tutorial" section of the driving test teaching platform and syncs it to the instructor's social media accounts, enhancing the reach of the instructional content. This multi-view instructional approach allows students to not only see the instructor's actual actions but also understand the route from a map perspective, making it easier to master the test route and key maneuvers. Combined with 3D map tracks, this helps students better understand road structure, driving tracks, and key test points, improving learning efficiency. Instructors no longer need to manually record additional map commentary; the system automatically generates visual tracks, enhancing the professionalism of the instructional videos. AI automatically analyzes and extracts location information and generates multi-view content, enabling instructors to quickly and efficiently produce high-quality instructional videos in batches, reducing their workload.

[0056] In one embodiment, it further includes:

[0057] In the case where it is determined through analysis that the video data includes external teaching video data, extracting timestamp information from metadata of the external teaching video data;

[0058] A time mapping is established between the captured driving trajectory and the external teaching video data based on the timestamp information, so that the captured driving trajectory in the multi-view teaching video data and the video frames in the external teaching video data are displayed correspondingly based on the time mapping.

[0059] Understandably, in the production of driving test instructional videos, simply displaying the driving trajectory may not be sufficient to help students accurately understand driving maneuvers. This solution extracts timestamp information from off-road instructional videos and synchronously maps the captured driving trajectory with the video content based on the timestamp. This ensures that when students watch the video, they can not only see the actual driving scene, but also simultaneously view the vehicle's position, turns, braking, and other key driving behaviors in the 3D map perspective at the corresponding time point. This time mapping technology makes multi-perspective instruction more accurate and helps students more clearly understand the relationship between driving maneuvers and the driving environment.

[0060] Exemplarily, when the coach uploads the off-road teaching video, the system first analyzes the video metadata to extract the timestamp information. Most devices such as dash cams, mobile phones, and GoPro recorders automatically embed timestamps when recording videos, which are used to mark the shooting time of each frame. By reading this timestamp information, the system can obtain the exact time corresponding to each frame of the video, ensuring that the subsequent trajectory data can be accurately matched with the video frames. For example, the coach uploads a video of the subject three exam route. The system reads the metadata of this video and obtains: start timestamp: 2025-03-12 14:30:05; end timestamp: 2025-03-12 14:45:20; and the total video duration is 15 minutes and 15 seconds, and the time interval between each frame is 0.033 seconds (30fps). The system will record the time information corresponding to all video frames to prepare for subsequent trajectory synchronization.

[0061] Exemplarily, in addition to the video timestamp, the system will also extract the timestamp information of the driving trajectory from the GPS data of the shooting device or external navigation devices (such as OBD vehicle devices, smart dash cams). Usually, the driving trajectory data contains multiple key fields including: timestamp, which records the GPS position of the vehicle at a certain time point; GPS coordinates (latitude, longitude), the location where the vehicle is at that time; speed, the driving speed of the vehicle at that time point; driving direction (heading), the orientation (angle) of the vehicle at that time. Then, based on the timestamp information, the system will perform time mapping between the driving trajectory data and the video frames to ensure that the trajectory information and the video content are played synchronously. The method of time mapping usually uses linear interpolation or time series matching algorithms (DTW, Dynamic Time Warping) to align the GPS data with the video frames. For example, the coach's vehicle passes through a traffic light intersection at 2025-03-12 14:35:10. The system will automatically match the video frame at this time point and ensure that the trajectory in the map shows that the vehicle just arrives at this position.

[0062] Exemplarily, after establishing the time mapping, the system will fuse the timestamp-synchronized driving trajectory data with the off-road teaching video to generate a multi-perspective teaching video. The main methods may include: Picture-in-Picture (PIP) mode, adding a small window (such as the lower right corner) to the video screen to display the vehicle driving trajectory on the 3D map and ensuring that the trajectory moves synchronously with the video screen. While watching the driving perspective of the coach, the trainee can see the trajectory position corresponding to the current time point and understand how operations such as vehicle speed and steering affect the driving path; Split-screen mode, with the driving video taken by the coach displayed on the left side of the video and the vehicle trajectory in the 3D map displayed on the right side. As the video plays, the vehicle icon on the map moves synchronously according to the timestamp to ensure that the information on both sides corresponds; Trajectory overlay mode, directly overlaying a semi-transparent driving trajectory on the original video screen and marking key positions (such as test points, lane change positions, braking points, etc.). When watching the driving operation, the trainee can intuitively understand the current position and the future driving path. For example, when the video plays to 2025-03-12 14:38:15, the coach is passing through a complex intersection: the in-vehicle perspective is shown on the left side of the screen: the coach steps on the brake, turns on the turn signal, and waits for the traffic light. The map perspective is shown on the right side of the screen: the vehicle trajectory icon stops at the intersection, and the traffic light animation synchronously shows the countdown to prompt the trainee to observe the change of the traffic signal.

[0063] Exemplarily, after generating the multi-perspective teaching video, the system will provide a preview interface for the coach to confirm the accuracy of the trajectory synchronization. If the coach believes that there is a deviation between the trajectory and the video screen, the accuracy of the time mapping can be adjusted manually. For example, adjust the time offset. If the trajectory is slightly delayed or advanced, the trajectory data can be fine-tuned to align it perfectly with the video frame; Select the trajectory visualization method. The coach can choose whether to use the Picture-in-Picture mode, split-screen mode, or trajectory overlay mode; Add voice explanations. The coach can record additional voice commentaries. For example: "The current location is the U-turn intersection during the test. Start after the traffic light turns green." For example, after the coach confirms the video and clicks "Confirm and Publish", the system will automatically upload the multi-perspective teaching video to the driving test teaching platform and synchronously share it on the social platform to maximize the teaching influence. Thus, through the time mapping technology, the trainee can clearly see the actual driving position, test route, and driving operations of the vehicle at different time points, which helps to improve spatial cognition and the passing rate of the test. Combining the real in-vehicle perspective with the map trajectory perspective provides a more intuitive learning method, and the trainee can understand driving skills from multiple angles. The coach does not need to manually align the trajectory, and the AI system will automatically extract the timestamp and establish a synchronization relationship, greatly saving the video production time. Automatically mark the key points of the test route (such as "Lane change area", "U-turn point", "Parking point") in the trajectory to help the trainee master the key points of the test.

[0064] In one embodiment, it further includes:

[0065] Query the real - scene map database based on the position parameter, and generate real - scene driving simulation data according to the captured driving trajectory in combination with the real - scene map database;

[0066] Fuse the real - scene driving simulation data into the picture of the off - road teaching video data to generate multi - perspective teaching video data.

[0067] It can be understood that, based on the off - road teaching video, the real - scene map database can be further combined to generate real - scene driving simulation data, and fuse it with the original teaching video to provide a more intuitive and realistic multi - perspective teaching experience. Compared with the traditional map trajectory display method, using the real - scene map can directly present information such as the driving environment, road signs, and buildings, enabling trainees to more clearly understand driving operations and the exam route. The GPS coordinates can be parsed from the data of the off - road teaching video to obtain the driving trajectory of the vehicle. Query the real - scene map database using the GPS coordinates to obtain the road real - scene data of the corresponding location. Based on the real - scene map data and combined with the GPS trajectory of the vehicle, generate a dynamic real - scene driving path for synchronously displaying the driving process of the trainee. Fuse the real - scene driving simulation data with the original teaching video, so that the trainee can not only see the actual driving picture of the coach but also view the driving route on the real - scene map, improving the learning effect.

[0068] Exemplarily, when the coach uploads the off-road teaching video, the system first parses the metadata of the video and extracts the GPS coordinates, timestamp, and speed information therein. These data usually come from dashcams, smartphones, or in-vehicle navigation systems and can be used to accurately obtain the driving trajectory of the vehicle. For example, the video metadata uploaded by the coach contains: starting coordinates: 116.403874, 39.914889 (xx Street, xx City); ending coordinates: 116.404874, 39.918889 (xx Park); total driving time: 15 minutes; sampling frequency: 1 second per point. After the system parses it, a complete GPS trajectory can be obtained to mark the driving route of the vehicle. After obtaining the driving trajectory, the system will call the real-scene map database to obtain the real-scene images or 3D map data of the corresponding section. Through these data sources, the system can obtain detailed information such as the road environment, traffic signs, and surrounding buildings around the driving path of the vehicle and match the real-scene pictures of each section based on the timestamp. For example, when the vehicle is at 116.403874, 39.914889 (xxx Street), the system calls the xx real-scene map to query the real-time street view data of this location and obtains: road signs (such as traffic lights, speed limit signs); surrounding buildings; road conditions (whether there is construction, traffic flow, etc.); weather conditions (if the map API supports weather information, the sense of reality can be enhanced). After obtaining the real-scene map data, the system will combine the GPS trajectory of the vehicle to create a dynamic real-scene driving simulation. Combining the GPS trajectory, a dynamic driving route is drawn on the real-scene map, marking the starting point, ending point, key turning points, parking points, etc. Through the frame-by-frame synchronization method, it is ensured that the movement of the vehicle in the map is consistent with the video content. The street view mode is adopted to simulate the driving process from the third-person or first-person perspective to enhance the sense of immersion. A 3D driving environment is constructed through the map data, and the driving process of the vehicle is dynamically rendered in the system to make it more visual. Key points of the exam are dynamically marked on the real-scene map, such as: "Turn on the turn signal in advance when changing lanes here", "The examiner may ask you to pull over here", "Pay attention to the speed camera ahead". After completing the real-scene driving simulation data, the system will intelligently fuse it with the off-road teaching video to generate the final multi-perspective teaching video. The method can include augmented reality (AR) overlay. For example, a semi-transparent driving trajectory is directly overlaid on the teaching video, allowing the trainee to see the current trajectory and the upcoming driving path while watching the real driving video. It is applicable to the VR / AR teaching mode to provide a more immersive learning experience. After completing the multi-perspective teaching video, the system will provide a preview interface for the coach to check the trajectory synchronization situation and make personalized adjustments, such as adjusting the trajectory visualization method (adjusting transparency, arrow size). Selecting the street view mode or the 3D mode. Adding additional voice explanations (such as: "Change lanes in advance at the upcoming intersection"). After confirmation, the coach can upload it to the driving test teaching platform with one key and synchronously share it on social media to expand the teaching influence.Thus, combined with the real - scene map, it makes it easier for trainees to understand the relationship between driving operations and the examination environment and remember the examination route. The real - scene simulation + real teaching videos enable trainees to not only see the driving process but also view the real - scene route, improving learning efficiency. Automatically generating real - scene teaching videos enables coaches to produce high - quality courses faster and more efficiently, reducing the burden of manual annotation. Highlighting examination key points, such as U - turn points and lane - change points, in the real - scene map can improve the passing rate of trainees.

[0069] Please refer to Figure 2 , an embodiment of the driving test teaching content generation device in the embodiment of the present application may include:

[0070] An acquisition unit 201, configured to acquire historical teaching video data and graphic and text content data published by a target coach user on the driving test teaching platform;

[0071] A generation unit 202, configured to generate target teaching video data based on the historical teaching video data and graphic and text content data through an AI large - model, and the target teaching video data is associated with the styles of the historical teaching video data and graphic and text content data of the target coach user;

[0072] A publishing unit, configured to upload the generated target teaching video data to a social platform bound to the target coach user and publish it on the social platform using the user account associated with the target coach user on the social platform.

[0073] In summary, the driving test teaching content generation device provided in the above - mentioned embodiment acquires historical teaching video data and graphic and text content data published by a target coach user on the driving test teaching platform; generates target teaching video data based on the historical teaching video data and graphic and text content data through an AI large - model, and the target teaching video data is associated with the styles of the historical teaching video data and graphic and text content data of the target coach user; uploads the generated target teaching video data to a social platform bound to the target coach user and publishes it on the social platform using the user account associated with the target coach user on the social platform. Thus, AI automatically generates teaching content, and coaches do not need to manually edit, dub, and edit, saving a large amount of time and improving teaching efficiency. AI learns the language habits and teaching logics of coaches, making the generated content highly consistent with the coach's style and enhancing brand influence. Compared with traditional manual video production, AI can quickly generate and regularly publish, improving the content activity of the platform. Combining technologies such as 3D animation, automatic subtitles, and voice explanations makes teaching videos more intuitive and easy to understand, improving the passing rate of trainees. Through automated multi - platform publishing, it expands the teaching coverage of coaches, attracts more trainee attention, and improves market competitiveness.

[0074] Above Figure 2The driving test teaching content generation device in the embodiments of the present application has been described from the perspective of modular functional entities. Next, the driving test teaching content generation device in the embodiments of the present application will be described in detail from the perspective of hardware processing. Please refer to Figure 3 , an embodiment of the driving test teaching content generation device 300 in the embodiments of the present application includes:

[0075] An input device 301, an output device 302, a processor 303, and a memory 304, where the number of processors 303 can be one or more, Figure 3 Taking one processor 303 as an example. In some embodiments of the present application, the input device 301, the output device 302, the processor 303, and the memory 304 can be connected by a bus or other means, where Figure 3 Taking connection by bus as an example.

[0076] Among them, by calling the operation instructions stored in the memory 304, the processor 303 is used to execute the above steps.

[0077] By calling the operation instructions stored in the memory 304, the processor 303 is also used to execute Figure 1 Any one of the corresponding embodiments.

[0078] Please refer to Figure 4 , Figure 4 This is a schematic diagram of an embodiment of the electronic system provided by the embodiments of the present application.

[0079] As Figure 4 shown, the embodiments of the present application provide an electronic system, including a memory 410, a processor 420, and a computer program 411 stored on the memory 420 and executable on the processor 420. When the processor 420 executes the computer program 411, the above steps are implemented.

[0080] In the specific implementation process, when the processor 420 executes the computer program 411, it can implement Figure 1 Any one of the corresponding embodiments.

[0081] Since the electronic system introduced in this embodiment is the device used to implement a driving test teaching content generation device in the embodiments of the present application, based on the method introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various variations of the electronic system in this embodiment. Therefore, the specific implementation of how this electronic system implements the method in the embodiments of the present application will not be described in detail here. As long as the device used by those skilled in the art to implement the method in the embodiments of the present application belongs to the scope of protection of the present application.

[0082] Please refer to Figure 5 , Figure 5Schematic diagram of an embodiment of a computer-readable storage medium provided by an embodiment of the present application.

[0083] As Figure 5 shown, this embodiment provides a computer-readable storage medium 500, on which a computer program 511 is stored. When the computer program 511 is executed by a processor, the above steps are implemented.

[0084] In the specific implementation process, when the computer program 511 is executed by a processor, it can implement Figure 1 any implementation manner in the corresponding embodiment.

[0085] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0086] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0087] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0088] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0089] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide for implementing the process Figure 1 in one process or multiple processes and / or boxes Figure 1 steps for the functions specified in one box or multiple boxes.

[0090] An embodiment of the present application also provides a computer program product, which includes computer software instructions. When the computer software instructions run on a processing device, the processing device is caused to execute a process such as Figure 1 the process in the driving test teaching content generation method in the corresponding embodiment.

[0091] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that a computer can store, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

[0092] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0093] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in electrical, mechanical, or other forms.

[0094] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0095] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0096] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. And the aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs and other various media that can store program codes.

[0097] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A method for generating driving test teaching content, characterized in that, Including: Obtain the historical teaching video data and graphic content data released by the target coach user on the driving test teaching platform; Generate target teaching video data based on the historical teaching video data and graphic content data through an AI large model, and the target teaching video data is associated with the styles of the historical teaching video data and graphic content data of the target coach user; Upload the generated target teaching video data to the social platform bound to the target coach user and publish it on the social platform using the user account associated with the target coach user on the social platform.

2. The method according to claim 1, wherein Also including: Obtain the video data uploaded by the target coach user to analyze the video data; When it is determined through analysis that the video data includes off-road teaching video data, extract the position parameters in the metadata of the off-road teaching video data; Query the three-dimensional map database based on the position parameters to obtain the corresponding shooting driving trajectory of the off-road teaching video data in the three-dimensional map; Fuse the shooting driving trajectory into the picture of the off-road teaching video data to generate multi-perspective teaching video data, and after the target coach user confirms, upload the multi-perspective teaching video data to the driving test teaching platform. The multi-perspective teaching video data includes the shooting perspective and the shooting driving trajectory perspective of the off-road teaching video data.

3. The method according to claim 2, wherein Also including: When it is determined through analysis that the video data includes off-road teaching video data, extract the timestamp information in the metadata of the off-road teaching video data; Fuse the shooting driving trajectory into the picture of the off-road teaching video data based on the timestamp information to generate multi-perspective teaching video data.

4. The method according to claim 3, wherein The step of fusing the shooting driving trajectory into the picture of the off-road teaching video data based on the timestamp information to generate multi-perspective teaching video data includes: Establish a time mapping between the shooting driving trajectory and the off-road teaching video data based on the timestamp information, so that the shooting driving trajectory in the multi-perspective teaching video data and the video frames in the off-road teaching video data are displayed correspondingly based on the time mapping.

5. The method according to claim 2, characterized in that, Also including: Query the real-scene map database based on the position parameters to generate real-scene driving simulation data according to the shooting driving trajectory in combination with the real-scene map database; Fuse the real-scene driving simulation data into the picture of the off-road teaching video data to generate multi-perspective teaching video data.

6. The method according to any one of claims 2 to 5, characterized in that The shooting driving trajectory is displayed in the multi-perspective teaching video data in the form of a window.

7. The method according to claim 5, wherein The real-scene driving simulation data is displayed in the multi-perspective teaching video data in the form of a window.

8. A driving test teaching content generation device, characterized in that, Including: An acquisition unit for obtaining the historical teaching video data and graphic content data released by the target coach user on the driving test teaching platform; A generation unit for generating target teaching video data based on the historical teaching video data and graphic content data through an AI large model, and the target teaching video data is associated with the styles of the historical teaching video data and graphic content data of the target coach user; A publishing unit, configured to upload the generated target teaching video data to a social platform bound to the target coach user and publish it on the social platform using the user account associated with the target coach user on the social platform.

9. An electronic system, comprising a memory and a processor, characterized in that, When the processor is used to execute the computer program stored in the memory, it implements the steps of the driving test teaching content generation method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the driving test teaching content generation method according to any one of claims 1 to 7.

Images