Refreshing symbol cross-modal interaction system

Abstract

The invention discloses a syllable symbol cross-modal interaction system which improves interaction quality and cultural fidelity by reconstructing a hardware synchronization mechanism, embedding a semantic coding library and fusing edge calculation and real-time rendering. By adopting the technical scheme of the invention, the method has the characteristics of multi-modal dynamic collaboration, cultural semantic fidelity and low-delay interaction.

Description

Technical Field

[0001] The present invention belongs to the field of human-computer interaction technology, and in particular relates to a mythological symbol cross-modal interaction system. Background Art

[0002] Mythological symbols are concrete projections of humanity's collective unconscious, and their evolution is essentially a microcosm of the struggle between power, belief, and technology. In his Mythologies, Roland Barthes proposed that mythology is a second-level symbolic system. First-level symbols consist of a "signifier" and a "signified," forming a direct "literal meaning." Mythology, on the other hand, is a network of sociocultural meanings (connotations) constructed through this basic symbolic system. Mythological symbols possess not only surface meaning but also implicit meaning and ideological function. Mythology conveys information through symbolic forms, and its core purpose is to reveal and illuminate the dominant values of a given period. The study of mythological semiotics has gradually shifted from a single cultural or historical perspective to an interdisciplinary and multidimensional exploration. At the 2024 International Conference on Semiotics and Visual Communication held in Cyprus, scholars engaged in an in-depth discussion of the concept of "contemporary myth," exploring the various manifestations of myth in modern society.

[0003] The mythological symbol system contains multimodal information such as visual patterns (such as taotie patterns and cloud-thunder patterns), auditory narratives (oral epics and sacrificial rituals), and tactile perception (such as casting marks on objects and material temperature), involving interdisciplinary knowledge such as semiotics, anthropology, art, and computer science. The digital protection and inheritance of mythological symbols has become an important topic in the integration of culture and technology. With the development of artificial intelligence and interactive technology, the digital application of mythological symbols has developed from single image scanning to multimodal interactions such as virtual reality (VR) and augmented reality (AR). However, existing technologies still have significant defects in multimodal collaboration, cultural semantic fidelity, and real-time interaction.

[0004] At present, the digital interactive system mainly adopts the following technical solutions:

[0005] 1. AR / VR single-modal display system

[0006] Data acquisition: Use 3D scanners to obtain surface geometric data of cultural relics (such as the point cloud model of Sanxingdui bronzes);

[0007] Visual rendering: Use Unity / Unreal engine to generate static texture maps;

[0008] Interaction design: Implement model rotation / zoom control based on gesture recognition (such as Leap Motion);

[0009] Audio Overlay: Pre-recorded audio commentary (triggered playback, not dynamically linked to the visual content).

[0010] Representative case: The Forbidden City AR Guide System (Patent CN112785647A): 3D models of cultural relics are displayed through AR glasses, and users control the perspective switching with gestures and click icons to play commentary audio.

[0011] 2. Tactile-visual local interaction devices

[0012] Tactile parameter mapping: extracting physical properties such as surface roughness and hardness of artifacts (e.g., vertex density of Egyptian stele meshes);

[0013] Haptic feedback generation: simulating material touch through the vibration of the motor of haptic gloves (HaptX Gloves);

[0014] Visual-tactile synchronization: The central processing unit (CPU) schedules rendering and tactile signals in a time-sharing manner.

[0015] Representative case: British Museum Haptic VR Project (IEEE VR 2021): When the user touches the surface of the virtual artifact, the tactile gloves trigger vibration feedback based on the grid density.

[0016] 3. Multimodal Cultural Heritage Database

[0017] Data storage in separate databases: images, text, and audio are stored independently in MySQL databases (such as the Dunhuang mural image database and the Sanskrit audio database);

[0018] Representative Cases:

[0019] Dunhuang Mural Multimodal Database (Patent Publication No. CN113515764A): Supports split-screen viewing of mural images and restoration records, but has no interactive function.

[0020] The defects of the existing technology and their causes are as follows:

[0021] 1. Multimodal fragmentation: sensory channel coordination failure

[0022] Defects: The visual, auditory, and tactile modules operate independently, lacking a dynamic coordination mechanism. For example, when a user touches a virtual bronze artifact, the tactile feedback is delayed (>50ms), causing it to be out of sync with the surface texture animation. The mythological narrative audio plays only on a timeline, failing to adjust content based on the user's gaze focus (e.g., the sacred tree pattern).

[0023] Cause: The hardware adopts a layered architecture (AR glasses, tactile gloves, and speakers are separate), and the software relies on the CPU to process signals serially, resulting in disordered multi-modal output timing.

[0024] 2. Cultural semantic distortion: insufficient fidelity in symbolic translation

[0025] Flaws: Existing technologies only reproduce the physical properties of symbols and fail to convey cultural metaphors. For example, the tactile feedback of the Taotie pattern is merely a "convex and concave" feeling, which fails to connect to the symbolic meaning of "suppressing evil spirits and exorcising evil spirits" in the Classic of Mountains and Seas. Similarly, the audio of the Sichuan River whistle is difficult to reconstruct if it is separated from the water velocity parameter model, revealing its connection to the flood control myth.

[0026] Cause: Parameter mapping relies on a single physical property (such as mesh vertex density) and fails to construct a cultural semantic encoding rule library (pattern-narrative-tactile mapping relationship).

[0027] 3. Disconnection between interaction and data: Lack of dynamic response capabilities

[0028] Defects: User behavior (such as gesture strength and eye movement) cannot drive multimodal feedback in real time. For example, a twisting gesture cannot dynamically adjust the density of symbol particle effects; and the database split-screen display mode prevents images, text, and audio from updating in tandem.

[0029] Cause: The database uses partitioned table storage and static retrieval logic, does not integrate a real-time rendering engine and sensor interface, and the interaction layer and data layer are completely separated.

[0030] Consequences of defects

[0031] (1) User experience degradation: Multimodal perception is disrupted and cultural semantics are distorted, resulting in a decrease in user immersion and cognitive depth (experience duration < 5 minutes); (2) Loss of cultural heritage value: (3) The philosophical connotation and artistic value of symbols are simplified into physical parameters during the digitalization process, weakening the effectiveness of cultural inheritance; Technical application limitations: The existing system is only suitable for simple scenarios such as museum tours and is difficult to support deep interactive needs such as education and the metaverse.

[0032] Existing technologies have structural defects in core links such as multimodal collaborative architecture, cultural semantic encoding rules, and real-time interaction engines, resulting in fragmented, low-fidelity, and high-latency digital interactions of mythological symbols. Summary of the Invention

[0033] The technical problem to be solved by the present invention is to provide a mythological symbol cross-modal interaction system with multimodal dynamic collaboration, cultural semantic fidelity, and low-latency interaction.

[0034] To achieve the above object, the present invention adopts the following technical solutions:

[0035] A cross-modal interaction system for mythological symbols comprises a multimodal sensor array, an edge computing terminal, and a feedback integration device. The edge computing terminal implements immersive interaction with mythological symbols through the collaborative design of a cultural semantic coding rule library and a dynamic mapping engine, based on the visual, tactile, and auditory sensory information of the multimodal sensor array and the visual feedback, tactile drive, and audio synthesis of the feedback integration device.

[0036] Preferably, the multimodal sensing array includes: a visual sensing module, a tactile sensing module, and an auditory sensing module; wherein the visual sensing module includes: AR glasses with an integrated 4K camera and an eye tracker; the tactile sensing module includes: a flexible pressure sensor array and a piezoelectric ceramic actuator; and the auditory sensing module includes: an 8-channel directional microphone array.

[0037] Preferably, the flexible pressure sensor array is embedded in the fingertips of the tactile glove.

[0038] Preferably, the piezoelectric ceramic actuators are arranged in a honeycomb pattern on the palm of the glove.

[0039] Preferably, the feedback integration device uses a Micro-OLED binocular display to achieve visual feedback through an HDMI 2.1 connection; uses an H-bridge chip to generate a 0-100V PWM signal to drive piezoelectric ceramics to achieve tactile drive; and uses a DSP processor to build a 3D spatial sound field to achieve audio synthesis.

[0040] Preferably, the edge computing terminal extracts the fractal dimension of the pattern based on the ResNet-50 model to establish a pattern-narrative mapping table, corresponds the user's gaze duration to the visual weight, and maps the touch pressure to the tactile frequency, thereby constructing a cultural semantic coding rule library.

[0041] Preferably, the edge computing terminal realizes visual generation and audio synthesis through a multimodal dynamic mapping engine and a real-time rendering engine based on a cultural semantic coding rule library.

[0042] Through the collaborative design of hardware heterogeneous architecture, cultural semantic coding rule library and dynamic mapping engine, the present invention systematically solves the multimodal fragmentation, semantic distortion and high latency problems of existing technologies, and provides complete technical support for immersive interaction of mythological symbols. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0044] Figure 1Schematic diagram of the structure of the mythological symbol cross-modal interaction system according to an embodiment of the present invention;

[0045] Figure 2 Flowchart for building a cultural semantic coding rule base;

[0046] Figure 3 Flowchart for achieving visual generation and audio synthesis. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] Example 1:

[0050] like Figure 1 As shown, an embodiment of the present invention provides a cross-modal interaction system for mythological symbols, including: a multimodal sensor array, an edge computing terminal and a feedback integration device; the edge computing terminal realizes immersive interaction of mythological symbols through the collaborative design of a cultural semantic coding rule library and a dynamic mapping engine based on the visual, tactile and auditory sensor information of the multimodal sensor array and the visual feedback, tactile drive and audio synthesis of the feedback integration device.

[0051] As an implementation manner of an embodiment of the present invention, a multimodal sensing array includes: a visual sensing module, a tactile sensing module, and an auditory sensing module.

[0052] Furthermore, the visual sensing module includes:

[0053] AR glasses integrate a 4K camera (resolution 3840×2160@60Hz);

[0054] The eye tracker (sampling rate 120Hz) is connected to the main controller via the MIPI CSI-2 interface.

[0055] Furthermore, the tactile sensing module includes:

[0056] Flexible pressure sensor arrays (16×16 dot matrix, ±0.1N accuracy) are embedded in the fingertips of the tactile gloves;

[0057] Piezoelectric ceramic actuator (100 points / cm 2 ) are arranged in a honeycomb pattern on the palm of the glove.

[0058] Furthermore, the auditory sensing module includes:

[0059] 8-channel directional microphone array (20Hz-20kHz) and bone conduction headphones are networked.

[0060] As an implementation method of an embodiment of the present invention, the edge computing terminal includes: a main control module and a communication interface

[0061] Furthermore, the main control module includes:

[0062] ARM Cortex-A78 is responsible for semantic processing;

[0063] FPGA (Xilinx Zynq UltraScale+) specializes in real-time signal processing;

[0064] ARM+FPGA heterogeneous communication is achieved through the AXI bus.

[0065] Furthermore, the communication interface includes:

[0066] PCIe 4.0 bus transmits sensor data (bandwidth 16GT / s);

[0067] Wi-Fi 6E / Bluetooth 5.3 dual-mode wireless protocol

[0068] As an implementation of an embodiment of the present invention, a feedback integration device is used for visual feedback, tactile drive and audio synthesis.

[0069] Furthermore, visual feedback is provided by a binocular Micro-OLED display (2560×2560 per eye, 90Hz refresh rate) connected via HDMI 2.1. Haptic drivers are provided by an H-bridge chip (DRV2605) that generates a 0-100V PWM signal to drive piezoelectric ceramics. Audio synthesis is provided by a DSP processor (TI TAS6424) that creates a 3D spatial sound field.

[0070] As an implementation method of an embodiment of the present invention, the heterogeneous computing packaging structure, feedback device structure, multimodal sensor array layout and edge computing terminal circuit design of the present invention are as follows:

[0071] 1. Heterogeneous computing packaging structure

[0072] Chip packaging: The ARM Cortex-A78 processor and the Xilinx Zynq UltraScale+ FPGA are vertically interconnected via TSV (diameter 10μm±1μm), with an interposer thickness of 100μm±5μm.

[0073] Bus connection: ARM and FPGA are directly connected using the AXI-Stream bus (64-bit width, frequency 500MHz±10MHz), with a bus spacing of 0.2mm±0.02mm;

[0074] Pin layout: LPDDR5 memory particles are connected to the FPGA through a BGA-468 package (ball diameter 0.4mm±0.05mm), with a pin spacing of 0.8mm±0.1mm.

[0075] 2. Feedback device structure

[0076] Structural composition: Piezoelectric ceramic array: hexagonal honeycomb arrangement (side length 3mm±0.1mm), each unit embedded with PZT-5H piezoelectric ceramic sheet (diameter 2mm±0.05mm, thickness 0.5mm±0.02mm);

[0077] Circuit connection: The ceramic chip is connected in series to the SPI interface through a silver paste printed circuit (line width 50μm±5μm, spacing 100μm±10μm);

[0078] Mechanical fixation: Silicone substrate (hardness 30 Shore A ± 5) with a groove depth of 0.5 mm ± 0.05 mm and a polyimide insulation layer (thickness 25 μm ± 3 μm) on the surface.

[0079] 4. Multimodal sensor array layout

[0080] AR glasses module:

[0081] Optical structure: The 4K camera (OV48C) and eye tracker (Tobii Nano) share an N-BK7 glass dual-path prism (optical axis angle 15°±0.5°), with a prism size of 10mm×10mm×5mm;

[0082] Installation structure: Aluminum alloy clips (thickness 1.2mm±0.1mm) are fixed to both sides of the nose pads with M2 screws (spacing 8mm±0.5mm).

[0083] Haptic Glove Module:

[0084] Pressure sensor: 16 × 16 FSR array (3 mm ± 0.1 mm diameter), connected to the I2C bus via serpentine copper traces (18 μm ± 2 μm thickness, 0.3 mm ± 0.05 mm spacing);

[0085] Hinge structure: TPU knuckle hinge (rotation angle 0-90°±5°), built-in FPC cable (width 2mm±0.1mm) bypassing the hinge axis.

[0086] 4. Edge computing terminal circuit design

[0087] Power module:

[0088] Battery parallel connection: Two 18650 lithium batteries are connected in parallel through the PMIC chip (TPS65813), and the positive contact uses a gold-plated spring pin (diameter 1.5mm±0.05mm, length 5mm±0.1mm);

[0089] Hot switching circuit: The AO3400 MOSFET and SS34 Schottky diode form the switching circuit and are placed on the bottom layer of the PCB (5mm±0.5mm from the edge).

[0090] Haptic drive circuit:

[0091] H-bridge topology: DRV2605 chip drives AON7400 MOSFET, and the output is connected to PESD5V0S1BA ESD protection diode (package 0402);

[0092] PCB design: 6-layer board structure (alternating signal layer, power layer, and ground layer), line width / spacing 0.1mm±0.02mm.

[0093] As an implementation method of an embodiment of the present invention, the edge computing terminal extracts the pattern fractal dimension based on the ResNet-50 model to establish a pattern-narrative mapping table, the user's gaze duration corresponds to the visual weight, the touch pressure intensity maps to the tactile frequency, and constructs a cultural semantic coding rule library. Figure 2 As shown, specifically including:

[0094] Collect user multimodal input data

[0095] It includes visual input (camera captures line of sight and pattern), tactile input (touchpad obtains pressure), and behavioral data (gaze duration, etc.).

[0096] Image feature extraction and pattern analysis

[0097] The ResNet-50 model is used to extract features from image patterns and calculate their fractal dimensions and other geometric properties.

[0098] Constructing a pattern-narrative mapping table

[0099] A mapping relationship is established between the fractal dimension of the pattern and the elements of mythological narrative to form a "pattern-narrative mapping table".

[0100] Extracting user visual behavior weights

[0101] Assign visual attention weights based on the user's gaze duration on a certain area.

[0102] Analyzing tactile input characteristics

[0103] Mapping touch pressure to different haptic feedback frequencies enhances immersion.

[0104] Generate cultural semantic encoding rules

[0105] Image patterns, visual attention weights, and tactile feedback frequencies are integrated to generate unified cultural semantic coding rule entries.

[0106] Building a semantic encoding rule base

[0107] Archive all entries into a structured semantic coding rule base for subsequent dynamic mapping and interactive use

[0108] Furthermore, the edge computing terminal realizes visual generation and audio synthesis based on the cultural semantic coding rule base through a multimodal dynamic mapping engine and a real-time rendering engine. Figure 3 As shown, specifically including:

[0109] Input multimodal perception data: Collect user's voice, gestures, images and other perception information, and pre-process it through the edge computing terminal.

[0110] Cultural semantic recognition and coding: Use the cultural semantic coding rule library to perform semantic recognition and structured coding on the perceived data to extract the mythological cultural elements.

[0111] Build a semantic instruction set: Convert the encoding results into a multimodal semantic instruction set that can be called by the engine.

[0112] Multimodal dynamic mapping engine processing: According to the semantic instruction set, the corresponding visual element and audio element resources are called to perform cross-modal mapping matching.

[0113] Real-time rendering engine execution generation: The mapping results are input into the real-time rendering engine to generate specific visual images (such as mythological characters, scene animations) and audio output (such as dubbing, background music).

[0114] User-side interactive output: The generated visual and audio information is fed back to the user-side display / playback interface to achieve an immersive cultural interactive experience.

[0115] Example 1 (Bronze Interaction):

[0116] Interaction triggering stage

[0117] User touch pressure detection:

[0118] Fingertip pressure 3.2N (sensor accuracy ±0.1N)

[0119] Eye tracking locks on the center area of the Taotie pattern (coordinate error ±0.5°)

[0120] Data processing stage

[0121] Semantic matching:

[0122] Calculation of pattern fractal dimension (D=1.72±0.03)

[0123] Related to the text fragment of "Classic of Mountains and Seas" (matching degree 92%)

[0124] Parameter generation:

[0125] Tactile frequency 200Hz (corresponding to 3.2N pressure)

[0126] Particle density 500 particles / cm 2 (Visual weight 40%)

[0127] Multimodal feedback stage

[0128] Haptic Execution:

[0129] Piezoelectric ceramic array responds within 10ms (vibration error ±2Hz)

[0130] Visual Rendering:

[0131] Micro-OLED display particle animation (90FPS frame rate)

[0132] Audio Output:

[0133] 3D sound field generates tiger roar (dynamic range 110dB)

[0134] Technical indicator verification

[0135] Delay Control:

[0136] End-to-end latency <10ms (9.2ms for tactile channel and 8.7ms for visual channel)

[0137] Semantic Fidelity:

[0138] Pattern-narrative matching accuracy ≥ 92% (100 test samples)

[0139] Energy consumption performance:

[0140] Continuous working time ≥ 4 hours (multi-mode full load state)

[0141] The embodiment of the present invention meets the three core requirements of the mythological symbol interaction system in terms of real-time performance (<10ms delay), cultural fidelity (≥92% accuracy), and user experience (4 hours of battery life).

[0142] Through the collaborative design of hardware heterogeneous architecture, cultural semantic coding rule library and dynamic mapping engine, we systematically solve the multimodal fragmentation, semantic distortion and high latency problems of existing technologies, and provide complete technical support for immersive interaction of mythological symbols.

[0143] Compared with the prior art, the present invention has the following characteristics:

[0144] 1. A multimodal dynamic collaboration mechanism, leveraging a heterogeneous hardware architecture (ARM+FPGA) and a timestamp alignment algorithm, achieves sub-frame synchronization (latency <10ms) of visual, tactile, and auditory signals. This addresses the issues of tactile feedback delay (>50ms) and misalignment of visual / auditory output caused by time-sharing scheduling in traditional systems. When a user touches a virtual bronze artifact, the tactile vibration (driven by piezoelectric ceramics) is strictly synchronized with the pattern particle animation and mythological narrative audio, enhancing the sense of immersion.

[0145] 2. Improved cultural semantic fidelity: Through a cultural semantic encoding rule library and symbol feature extraction algorithm, accurate mapping of pattern, narrative, and sound effects is achieved (semantic matching accuracy ≥ 92%). Technical Effect: This solves the problem of missing cultural metaphors caused by traditional parameter mapping relying solely on physical properties (such as roughness). When the user gazes at the Taotie pattern on a bronze artifact, the system automatically associates it with the "suppressing evil spirits and driving away evil spirits" semantics from the Classic of Mountains and Seas, synchronously triggering a tiger roar sound effect (frequency band 80-200Hz) and dynamic flame particle effects, improving the efficiency of cultural information transmission.

[0146] 3. Optimize tactile interaction resolution

[0147] Through the honeycomb piezoelectric ceramic array (100 points / cm 2 ) and high-frequency drive circuit (50-500Hz) to achieve micron-level tactile feedback. Solve the low tactile resolution of traditional ERM motors (20 points / cm 2 ) and the problem of single feedback frequency (<100Hz). Users can feel the casting marks of bronze ware (0.1mm level bumps) and the warm touch of jade cloud and thunder patterns, and the tactile fidelity is improved.

[0148] 4. Real-time interaction and offline usability enhancement

[0149] Through edge computing terminals (localized processing) and dual-battery redundant power supply, smooth interaction is achieved in offline environments. This solves the high latency (>100ms) and offline unavailability issues caused by traditional cloud-based systems. In offline scenarios such as museums and archaeological sites, it still supports real-time rendering (latency <15ms) and long battery life, extending its applicability to outdoor cultural heritage protection scenarios.

[0150] Through the collaborative innovation of heterogeneous hardware acceleration, cultural semantic coding rule base and multimodal dynamic mapping engine, the defects of traditional technologies in multimodal collaboration, cultural fidelity, real-time interaction, etc. are systematically addressed, so that the digital interaction of mythological symbols can achieve a leap forward in immersion, cultural heritage effectiveness and breadth of applicable scenarios, providing core technical support for the dynamic protection and dissemination of cultural heritage.

[0151] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A mythological symbol cross-modal interaction system, characterized by: include: Multimodal sensor arrays, edge computing terminals, and feedback integration devices; The edge computing terminal realizes immersive interaction of mythological symbols through the collaborative design of the cultural semantic coding rule library and the dynamic mapping engine based on the visual feedback, tactile drive and audio synthesis of the feedback integration device of the multimodal sensor array and the visual, tactile and auditory sensor information and feedback integration device of the multimodal sensor array.

2. The mythological symbol cross-modal interaction system according to claim 1, characterized in that: The multimodal sensing array includes: visual sensing module, tactile sensing module, and auditory sensing module; among them, the visual sensing module includes: AR glasses with integrated 4K camera and eye tracker; the tactile sensing module includes: flexible pressure sensor array and piezoelectric ceramic actuator; the auditory sensing module includes: 8-channel directional microphone array.

3. The mythological symbol cross-modal interaction system according to claim 2, characterized in that: Flexible pressure sensor arrays embedded in the fingertips of tactile gloves.

4. The mythological symbol cross-modal interaction system according to claim 3, characterized in that: Piezoelectric ceramic actuators are arranged in a honeycomb pattern on the palm of the glove.

5. The mythological symbol cross-modal interaction system according to claim 4, characterized in that: The feedback integration device uses a Micro-OLED binocular display to achieve visual feedback through an HDMI 2.1 connection; an H-bridge chip is used to generate a 0-100V PWM signal to drive piezoelectric ceramics to achieve tactile drive; and a DSP processor is used to build a 3D spatial sound field to achieve audio synthesis.

6. The mythological symbol cross-modal interaction system according to claim 5, characterized in that: The edge computing terminal extracts the fractal dimension of the pattern based on the ResNet-50 model to establish a pattern-narrative mapping table, corresponds the user's gaze duration to the visual weight, and maps the touch pressure to the tactile frequency, thereby constructing a cultural semantic coding rule library.

7. The mythological symbol cross-modal interaction system according to claim 6, characterized in that: The edge computing terminal is based on a cultural semantic coding rule library and realizes visual generation and audio synthesis through a multimodal dynamic mapping engine and a real-time rendering engine.

Images