A virtual scene display system and method based on a rotating mechanism

Abstract

The application provides a virtual scene display system and method based on a rotating mechanism. It relates to the technical field of virtual reality and augmented reality display, and comprises a rotating drive assembly, an absolute value angle encoder and a bearing platform. The bearing platform is used for bearing an audience and rotating around a vertical shaft. A posture acquisition unit is used for outputting a rotating posture parameter in real time through the angle encoder. An audience perception-collaborative control unit comprises a plurality of depth cameras, which are used for collecting audience quantity and orientation data. The virtual scene display system and method based on the rotating mechanism have the advantages of high precision, low delay, easy expansion, multi-person immersion and operation-friendly, which are significantly superior to existing fixed screen or one-way rotating display technology.

Description

A virtual scene display system and method based on a rotating mechanism Technical Field

[0001] This invention relates to the field of virtual reality and augmented reality display technology, specifically to a virtual scene display system and method based on a rotating mechanism. Background Technology

[0002] Existing virtual scene display equipment can be broadly categorized into head-mounted displays, dome / circular projection systems, 360° panoramic LED screens, and "rotating booths + single-sided displays" devices. Head-mounted displays rely on posture sensors to provide immersive stereoscopic images for a single user; they are portable but difficult for multiple people to share. Circular theaters use multi-channel projection to stitch together ultra-wide field-of-view images on a fixed screen, serving groups of viewers simultaneously, but they are expensive, have strict site requirements, and limited interactivity. Panoramic LED or flexible OLED cylindrical screens surround the audience with a physical screen, offering strong visual impact, but the viewpoint remains fixed and content production costs are high. Some interactive exhibits slowly rotate the audience or objects, rendering multi-angle visual effects on a single large screen; however, this solution is mostly unidirectional mechanical rotation, lacking precise posture feedback and real-time image interaction.

[0003] The above solutions have revealed several shortcomings in application: head-mounted displays can easily cause dizziness with prolonged wear and are difficult to maintain in public settings; while large screens and panoramic screens support multiple viewers, fixed viewpoints lead to perspective distortion and a lack of a sense of presence and displacement; rotating booth solutions are prone to visual misalignment due to the asynchrony between rendering and mechanical movement, and rapid rotation also brings safety and wear issues; in addition, large projection / LED systems are costly to build and calibrate, and mechanical platforms are generally bulky and have low modularity, making it difficult to meet the urgent needs of cultural tourism exhibitions and other scenarios for low cost, easy deployment, and comfortable multi-person interaction. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a virtual scene display system and method based on a rotating mechanism, which solves the problem of how to achieve immersive panoramic display with multiple users synchronously shared by coupling high-precision attitude acquisition of a rotating platform with adaptive sector rendering of audience distribution.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a virtual scene display system and method based on a rotating mechanism, comprising:

[0006] A rotation drive assembly includes a support platform and an absolute angle encoder, the support platform being used to support spectators and rotate about a vertical axis;

[0007] An attitude acquisition unit is used to output rotational attitude parameters in real time through the angle encoder;

[0008] The audience perception-cooperative control unit includes several depth cameras, which are used to collect audience number and orientation data. The audience perception-cooperative control unit divides the ring display unit into several sectors based on the audience number and orientation data, and generates sector mapping parameters and display parameters.

[0009] The real-time rendering unit generates video signals corresponding to each sector based on the rotational attitude parameters and the sector mapping parameters.

[0010] The ring-shaped display unit is used to receive the video signal and drive each sector independently according to the display parameters, synchronously presenting a 360° virtual scene, so that multiple viewers can have an immersive viewing experience at the same time.

[0011] Preferably, the absolute angle encoder has an angle resolution of no more than 0.05° and an encoding frequency of no less than 1000Hz to ensure high-precision attitude feedback.

[0012] Preferably, the attitude acquisition unit uses an extended Kalman filter to eliminate mechanical jitter and completes attitude calculation within 5ms.

[0013] Preferably, the audience perception-cooperative control unit uses a density clustering algorithm to group the point cloud acquired by the depth camera, and dynamically determines the number of sectors of the ring display unit based on the average orientation of the audience groups, wherein the number of sectors N is in the range of 2≤N≤12.

[0014] Preferably, the real-time rendering unit calculates the perspective projection matrix and generates the video signal according to the following model formula:

[0015]

[0016] in, For sector i Projection matrix, For the camera intrinsic parameter matrix, Let be the rotation matrix about the vertical axis. The current platform rotation angle, For the first Azimuth of the sector center.

[0017] Preferably, the brightness of the ring-shaped display unit is determined using the following model formula. Adaptive adjustment:

[0018]

[0019] in, As the reference brightness, This is the brightness adjustment coefficient. Let i be the number of viewers in sector i. This represents the current total number of sectors.

[0020] Preferably, the virtual scene display system based on the rotating mechanism further includes a network communication unit, which uses 5G to realize the real-time transmission of video signals and display parameters between the real-time rendering unit and the ring display unit.

[0021] A virtual scene display method based on a rotating mechanism includes:

[0022] S1. Obtain the rotation angle of the platform around the vertical axis by an absolute angle encoder installed on the carrier platform, and output the denoised attitude parameters using an extended Kalman filter.

[0023] S2. Collect audience point cloud data using a depth camera, perform density clustering on the point cloud data, and obtain the audience location and number of audience members.

[0024] S3. Based on the audience's orientation, the ring-shaped display unit is divided into N sectors, and brightness adjustment parameters are generated according to the number of viewers in each sector, so that the sectors with more viewers receive higher brightness, thereby improving viewing balance.

[0025] S4. Call the 3D graphics engine to perform corresponding perspective transformations for each sector based on the posture parameters and generate video signals, ensuring that the system end-to-end latency is no more than 20ms.

[0026] S5. The video signal and the brightness parameters are sent synchronously to the ring display unit to output the virtual scene, so that multiple viewers can simultaneously enjoy a 360° immersive viewing experience without wearing head-mounted devices.

[0027] This invention provides a virtual scene display system and method based on a rotating mechanism. It has the following beneficial effects:

[0028] This virtual scene display system and method based on a rotating mechanism seamlessly couples the motion of the mechanical platform with the virtual image through a four-level closed loop: "high-precision rotational posture acquisition + adaptive sector division for audience distribution + real-time multi-viewport rendering + zoned brightness equalization". On the one hand, the combination of absolute angle encoder and extended Kalman filter controls the rotation error to within 0.05° and the end-to-end latency to ≤20ms, basically eliminating perspective distortion and dizziness. On the other hand, the depth camera array senses the audience's position in real time and dynamically matches the display content and brightness with 2 to 12 variable sectors. This avoids the poor experience of "clear center and dark edges" in large-screen solutions and allows the audience to obtain a 360° immersive field of view without wearing a head-mounted display, achieving the core goal of "multi-person sharing + comfortable viewing".

[0029] The supporting methodology streamlines the aforementioned hardware capabilities: first, it acquires and denoises the posture in real time; then, it automatically determines the number of sectors and their corresponding brightness based on audience clustering results; next, during the rendering stage, it outputs synchronized video by performing independent perspective transformation on each sector; and finally, it drives the circular screen in sections. This process simplifies deployment and debugging, the brightness adaptive mechanism ensures viewing balance under different audience densities, and 5G low-latency transmission balances high bandwidth and high reliability. Overall, this technical solution combines the advantages of high precision, low latency, easy scalability, multi-user immersion, and ease of operation and maintenance, significantly outperforming existing fixed-screen or unidirectional rotating display technologies. Attached Figure Description

[0030] Figure 1 is a schematic diagram of the process for realizing the invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] As shown in Figure 1, this embodiment of the invention provides a virtual scene display system and method based on a rotation mechanism, including a rotation drive component, comprising a support platform and an absolute angle encoder. The support platform is used to support the audience and rotate around a vertical axis. The absolute angle encoder has an angle resolution of no more than 0.05° and an encoding frequency of no less than 1000Hz to ensure high-precision attitude feedback.

[0033] The specific implementation method is as follows:

[0034] The platform utilizes a 2000mm diameter aluminum alloy honeycomb structure, weighing 85kg and designed with a static load capacity of 600kg, accommodating 6 to 8 adults simultaneously. The platform is rigidly connected to a permanent magnet synchronous servo motor and planetary reducer below via an integrated turntable base. The motor has a rated power of 2.2kW, a rated output torque of 35N·m, and a reduction ratio of 30:1, enabling uniform or variable speed rotation from 0 to 30rpm, with a peak angular acceleration of 60° / s².

[0035] Attitude detection utilizes a Heidenhain absolute grating encoder with a single-turn resolution of 21 bits, corresponding to an angular resolution of 0.00026° and a refresh rate of 1200Hz. The encoder employs a tapered sleeve positioning system, controlling coaxiality error within 0.02mm and radial runout to no more than 10µm. Signal transmission is achieved via a built-in fiber optic slip ring, enabling 360° continuous rotation without cable tangling.

[0036] The main controller reads encoder angle data at a frequency of 1000Hz and runs an extended Kalman filter, with an average end-to-end attitude calculation delay of 3.8ms. A constant speed test at 15rpm showed a static angle mean square error of 0.018°, a dynamic error of 0.032°, and a maximum instantaneous error of 0.041°. The system ran continuously for 24 hours without any data frame loss.

[0037] Fault protection logic: When the dynamic error is continuously greater than 0.05°, the speed will automatically decrease to 5 rpm and an alarm will be triggered; when the communication heartbeat is lost for more than 100 ms, braking will be performed with a braking distance of 12.5 mm.

[0038] In a demonstration environment that simultaneously drives four channels of 4K 60fps graphics rendering, the rotation drive component consistently maintains synchronization between the image and the actual viewpoint, without tearing or motion blur, verifying the reliability of its high-precision attitude feedback and long-term stable operation.

[0039] The attitude acquisition unit is used to output rotational attitude parameters in real time via an angle encoder. The attitude acquisition unit employs an extended Kalman filter to eliminate mechanical jitter and completes attitude calculation within 5ms.

[0040] The audience perception-cooperative control unit includes several depth cameras, which are used to collect audience number and orientation data. Based on the audience number and orientation data, the audience perception-cooperative control unit divides the ring-shaped display unit into several sectors and generates sector mapping parameters and display parameters.

[0041] The audience perception-cooperative control unit uses a density clustering algorithm to group the point cloud acquired by the depth camera and dynamically determines the number of sectors of the ring display unit based on the average orientation of the audience groups. The number of sectors N ranges from 2 to N to 12.

[0042] The specific implementation method is as follows:

[0043] Twelve depth cameras are evenly spaced along the outer edge of the ring-shaped display unit, with a horizontal field of view of 87° and a vertical field of view of 58°. Each camera is installed at a height of 2.3m, with an inward tilt angle of 15°, covering an area with a radius of 6m from the audience's standing area to the central platform. The cameras are powered and transmit data via Gigabit Ethernet, with a depth frame resolution of 1280×720 and a frame rate of 60Hz. All raw depth streams are aggregated to an edge server, with the server's end-to-end processing latency kept below 35ms.

[0044] The server first converts the 12-channel depth map into point cloud data and performs downsampling at a voxel size of 0.1m to reduce computational load. Then, it calls a density-based clustering algorithm, setting the Euclidean distance threshold to 0.4m and the minimum cluster size to 5 pixels, to identify 3D point cloud clusters of standing audience members in real time. The algorithm outputs the centroid coordinates of each cluster and converts them into an azimuth angle φ and radius r relative to the center stage, while simultaneously counting the number of depth points n contained in that cluster as an estimate of the number of audience members. The entire clustering-conversion process runs every 100ms.

[0045] The system, based on the total number M of audience clusters identified at the current moment, merges clusters with an azimuth difference of less than 12° according to the "nearest merging" principle, resulting in the final number of audience clusters N, ensuring that N is between 2 and 12. Then, it allocates N sectors to the ring-shaped display unit in a clockwise direction, with the sector centerline coinciding with the azimuth of each audience cluster. The angle of a single sector is calculated by dividing 360° by N. If the number of audiences corresponding to a certain sector is greater than the global average number of audiences, the system improves the visibility of that sector through brightness enhancement and gamma fine-tuning, with the enhancement amount linearly mapped to within a 20% range. All sector mapping parameters and brightness parameters are packaged and sent to the real-time rendering unit in each update cycle.

[0046] For example, if there are 30 audience members present, and the clustering algorithm identifies 8 audience clusters, the system dynamically divides the audience into 8 sectors, each with a 45° angle. The third sector has 6 audience members, 3.75 more than the average, and its brightness is increased by 15%. The entire computation and broadcast link latency is maintained within 50ms, ensuring that sector division and brightness adjustment are almost synchronized with the actual audience distribution, thus achieving balanced optimization of the immersive viewing experience for multiple audience members.

[0047] The real-time rendering unit generates video signals corresponding to each sector based on rotational attitude parameters and sector mapping parameters. The real-time rendering unit calculates the perspective projection matrix and generates the video signal according to the following model formula:

[0048]

[0049] in, For sector i Projection matrix, For the camera intrinsic parameter matrix, Let be the rotation matrix about the vertical axis. The current platform rotation angle, For the first Azimuth of the sector center.

[0050] The circular display unit receives video signals and independently drives each sector according to display parameters, synchronously presenting a 360° virtual scene, allowing multiple viewers to enjoy an immersive viewing experience simultaneously. The circular display unit uses the following model formula to control brightness. Adaptive adjustment:

[0051]

[0052] in, As the reference brightness, This is the brightness adjustment coefficient. Let i be the number of viewers in sector i. This represents the total number of sectors currently. Example 1

[0053] The specific implementation method is as follows:

[0054] Application scenario: Conventional exhibition mode of digital museums.

[0055] An 8-sector panoramic screen system was deployed in the second-floor rotunda of the Shenzhen Futian Digital Museum. The supporting platform has a diameter of 2000mm and a thickness of 120mm, constructed with aluminum alloy honeycomb panels and a carbon fiber shell. The platform itself weighs 85kg, with a designed static load of 600kg, and can accommodate eight audience members simultaneously. The rotation drive is powered by a 2.2kW permanent magnet synchronous servo motor paired with a 30:1 planetary reducer, achieving a maximum speed of 10rpm, an angular acceleration of 60° / s², and a positioning error controlled within 0.04°. Attitude data is output at a frequency of 1200Hz, processed by an extended Kalman filter with an end-to-end delay of 3.8ms.

[0056] The real-time rendering unit consists of two dual-socket Xeon servers, each equipped with four RTX 4090 graphics cards, providing a total graphics computing power of approximately 950 TFLOPS. The servers are interconnected with the attitude control and display nodes via dual 10 Gigabit Ethernet connections. The rendering pipeline is based on Unreal Engine 5: each frame reads the latest attitude angle and the center positions of eight sectors, dynamically adjusts the camera matrix, and outputs eight channels of 4K 60fps video; the system's end-to-end latency is consistently maintained at 18ms.

[0057] The panoramic screen uses curved LED modules with a 1.5mm pitch, with a total circumference of 30m and a height of 3.6m; each sector has a resolution of 3840×2160. Twelve Intel RealSense D435 depth cameras are evenly spaced along the screen's eaves, capturing point clouds at 60Hz. Edge servers use the DBSCAN algorithm to cluster audience groups in real time, with an average computation time of 12ms. The system reassesses the audience distribution every 200ms; when the number of audience members in a sector exceeds the overall average by more than 30%, the brightness of that sector is increased by 15%, while sparsely populated sectors are reduced by 5%, with overall power consumption kept below 700W.

[0058] After 30 days of operation, the platform has accumulated 700 hours of rotation and received an average of 3,500 visitors per day. A satisfaction survey showed that 97% of viewers found the experience highly immersive and reported no significant dizziness. Aside from daily screen cleaning and weekly encoder bolt tightening, there are no additional maintenance costs.

[0059] A virtual scene display method based on a rotating mechanism includes:

[0060] S1. Obtain the rotation angle of the platform around the vertical axis by an absolute angle encoder installed on the carrier platform, and output the denoised attitude parameters using an extended Kalman filter.

[0061] S2. Collect audience point cloud data using a depth camera, perform density clustering on the point cloud data, and obtain the audience location and number of audience members.

[0062] S3. Based on the audience's location, the ring-shaped display unit is divided into N sectors, and brightness adjustment parameters are generated according to the number of viewers in each sector, so that the sectors with more viewers receive higher brightness, thereby improving viewing balance.

[0063] S4. Call the 3D graphics engine to perform corresponding perspective transformations for each sector based on the pose parameters and generate video signals, ensuring that the system end-to-end latency is no more than 20ms.

[0064] S5. The video signal and brightness parameters are simultaneously sent to the circular display unit to output the virtual scene image, allowing multiple viewers to simultaneously enjoy a 360° immersive viewing experience without needing to wear head-mounted displays. Example 2

[0065] Unlike Example 1, the application scenario of this example is a high-load impact mode for theme parks.

[0066] An immersive amusement park in Pudong, Shanghai, has created a 600m² sci-fi shuttle project on its basement level, with a 12-sector ultra-high-definition panoramic screen at its core. The support platform has a diameter of 3000mm, a titanium alloy frame with tempered glass steps, a platform weight of 210kg, and a designed static load of 1000kg, accommodating 15 audience members. It is driven by a 5kW water-cooled direct-drive motor, accelerating from 0 to 25rpm in 12 seconds. The absolute encoder has a refresh rate of 1400Hz and a resolution of 0.00026°, forming a redundant attitude chain with dual inertial measurement units. If the difference between the two data streams exceeds 0.05°, deceleration is triggered.

[0067] The rendering cluster consists of four servers equipped with NVIDIA A100 processors, totaling 160GB of video memory, capable of real-time output of 12 streams of 8K 90fps video. Pose and sector mapping parameters are broadcast via a 25GbE switch, with an end-to-end latency of 15ms from the rendering node to the display. The engine uses distributed deep buffering technology to eliminate sector seam flicker.

[0068] The panoramic screen uses 1.2mm pitch Micro-LED modules with a total pixel count exceeding 300 million. The system features adaptive brightness: when the crowd density in a single sector exceeds 4 people / m², the brightness increases to a peak of 120%, and decreases to 80% when the density is below 1 person / m², with overall power consumption controlled below 20kW. To complement the dynamic floor, vibration sensors provide real-time acceleration feedback to the rendering engine, dynamically adjusting the camera's near-cropping plane to prevent image jitter.

[0069] During a continuous 72-hour stress test, the platform rotation and image synchronization error was a maximum of 0.018°, and the brightness difference across the 12 sectors remained within 3%. At peak operating hours, 1450 people were served per hour, ensuring a consistent visual experience for every viewer. Nighttime maintenance only required replacing the UPS module and cleaning the camera lenses, taking a total of 30 minutes to restore operation. Example 3

[0070] Unlike the previous example, this example is used in an immersive display mode for a company's new car launch hall.

[0071] A six-sector panoramic screen system was deployed in the multimedia launch hall of a car brand headquarters in Beijing for new model launches and media experiences. The venue is a circular space with a diameter of 14m, featuring a central rotating platform with a diameter of 2500mm. The platform's platform platform is made of carbon fiber honeycomb composite material, weighs 120kg, and has a designed static load of 800kg. It can accommodate 10 people simultaneously, including senior company executives, the design team, and the presenter. The platform is driven by a 3kW water-cooled direct-drive servo motor with a reduction ratio of 25:1, a maximum speed of 12rpm, and an angular acceleration of 45° / s². Attitude data is output at a frequency of 1200Hz, with an extended Kalman filter and an end-to-end delay of 4ms.

[0072] The panoramic screen uses curved LED modules with a 1.6mm pitch, measuring 22m in length and 3m in height, and comprises six sectors, each with a resolution of 3840×1920, capable of fully displaying close-ups of vehicle exteriors, interior perspectives, and dynamic road conditions. Eight Intel RealSense L515 laser depth cameras are mounted on the top, capturing audience point clouds at 60Hz. Edge servers use density clustering to identify audience clusters and estimate the number of people in real time; when the number of media personnel in a single sector exceeds 20% of the overall average, the system increases the brightness of that sector by 10% and decreases the brightness of other sectors by 5% to guide camera and video focus.

[0073] The real-time rendering cluster consists of three dual-socket Xeon servers, each equipped with two RTX 6000 graphics cards, capable of simultaneously outputting six streams of 5K 75fps video. Platform attitude angles and sector mapping parameters are broadcast via a 10GbE switch, with end-to-end latency from rendering to the display controlled within 22ms. To demonstrate vehicle color accuracy, the system performs color lookup table correction before the end of each frame rendering, keeping the LED screen's ΔE95 within 1.5.

[0074] To facilitate the physical vehicle's entry, the platform can accelerate from 0 rpm to a stable display speed of 8 rpm within 60 seconds; on-site laser line scanners ensure a minimum safe distance of 400mm between the vehicle's outline and the panoramic screen. At its peak, the launch event hosted 300 media and guests in a single session, with the system running continuously for 5 hours without any frame drops or synchronization anomalies. Visitor surveys showed that 92% of media representatives believed the panoramic display effectively enhanced the immersive experience and visual impact of showcasing the vehicle's highlights.

[0075] Routine maintenance only requires wiping the LED screen every 2 days, checking the cleanliness of the camera lenses, and calibrating the platform encoder zero position once. The average monthly downtime for maintenance is less than 40 minutes, which meets the operational requirements of enterprises with high-frequency product releases.

[0076] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A virtual scene display system based on a rotating mechanism, characterized in that, include: A rotation drive assembly includes a support platform and an absolute angle encoder, the support platform being used to support spectators and rotate about a vertical axis; The attitude acquisition unit is used to output rotational attitude parameters in real time through the angle encoder; the audience perception-cooperative control unit includes several depth cameras, which are used to collect audience number and orientation data. The audience perception-cooperative control unit divides the ring display unit into several sectors according to the audience number and orientation data, and generates sector mapping parameters and display parameters. The real-time rendering unit generates video signals corresponding to each sector based on the rotational attitude parameters and the sector mapping parameters. The ring-shaped display unit receives the video signal and independently drives each sector according to the display parameters, synchronously presenting a 360° virtual scene; the real-time rendering unit calculates the perspective projection matrix and generates the video signal according to the following model formula: in, For sector i Projection matrix, For the camera intrinsic parameter matrix, Let be the rotation matrix about the vertical axis. The current platform rotation angle, For the first Sector center azimuth angle; the brightness of the ring-shaped display unit is determined by the following model formula. Adaptive adjustment: ;in, As the reference brightness, This is the brightness adjustment coefficient. Let i be the number of viewers in sector i. This represents the current total number of sectors.

2. The virtual scene display system based on a rotating mechanism according to claim 1, characterized in that: The absolute angle encoder has an angular resolution of no more than 0.05° and an encoding frequency of no less than 1000Hz.

3. The virtual scene display system based on a rotating mechanism according to claim 1, characterized in that: The attitude acquisition unit uses an extended Kalman filter to eliminate mechanical jitter and completes attitude calculation within 5ms.

4. The virtual scene display system based on a rotating mechanism according to claim 1, characterized in that: The audience perception-cooperative control unit uses a density clustering algorithm to group the point cloud acquired by the depth camera, and dynamically determines the number of sectors of the ring display unit based on the average orientation of the audience groups. The number of sectors N is in the range of 2≤N≤12.

5. A virtual scene display system based on a rotating mechanism according to claim 1, characterized in that: The virtual scene display system based on the rotating mechanism also includes a network communication unit, which uses 5G to realize the real-time transmission of video signals and display parameters between the real-time rendering unit and the ring display unit.

Images