A rich range of authentication systems, sender and receiver

Abstract

This application provides an authentication system, a sender, and a receiver with rich detail range. The authentication system includes a sender and a receiver connected by communication. The receiver performs video detail enhancement processing according to a detail rendering protocol through a content processing module. An effect evaluation module evaluates the processing effect and generates an evaluation result. A credential generation module collects relevant data and combines it with the protocol and evaluation result to generate an authentication credential, which is then sent to the sender. The sender's verification module verifies the authentication credential and evaluation result based on the detail rendering protocol. This invention achieves standardized processing and verifiable authentication of detail enhancement effects, reduces presentation differences between heterogeneous devices, and ensures the consistency and reliability of video content detail presentation.

Description

Technical Field

[0001] This application relates to the field of display effect evaluation technology, and in particular to a richly detailed authentication system, a transmitter, and a receiver. Background Technology

[0002] In modern video content distribution and display scenarios, bandwidth limitations, the limitations of compression algorithms, and the performance differences of heterogeneous devices often lead to problems such as loss of spatial detail, blurred textures, and soft edges during video transmission. Especially in streaming media and ad-supported distribution environments, low- or medium-bitrate content, after transmission, not only suffers significant compression of spatial detail but is also prone to compression artifacts such as blocky distortion, ringing, and mosquito noise. As consumers increasingly use large-format, high-resolution, and high-refresh-rate displays to watch videos, these image quality defects are further amplified, severely impacting the user's viewing experience.

[0003] To address these issues, the industry employs various video enhancement techniques, such as super-resolution to increase resolution, noise reduction algorithms to suppress noise, sharpening to enhance edge clarity, and motion estimation and compensation to optimize motion smoothness. However, these enhancement techniques are typically deployed as independent modules, lacking effective collaborative control mechanisms, leading to conflicting optimization effects. For instance, while over-sharpening can enhance edge details, it may introduce halos and ringing artifacts; temporal filtering can reduce flickering but blurs fine textures; and motion interpolation, while smoothing motion, may amplify inconsistencies in inter-frame details, ultimately reducing the overall image quality's coherence.

[0004] Furthermore, existing video enhancement solutions lack a unified standard framework. They lack clear definitions of the scope of detail and quantitative measurement methods, as well as an effective negotiation mechanism between device capabilities and content requirements. The parameter settings of existing solutions largely rely on fixed rules or heuristic strategies, failing to adaptively adjust to dynamic changes in content characteristics, network conditions, and device capabilities. This results in significant differences in the presentation of the same content across different devices. Simultaneously, the lack of verifiable authentication mechanisms makes it difficult to objectively assess the quality of enhancement effects, failing to provide reliable quality assurance for content distributors, device manufacturers, and users. This hinders the achievement of a consistent, detailed presentation experience across devices and scenarios, severely restricting the industrialization and ecosystem collaboration of video display technology. Summary of the Invention

[0005] This application provides a rich detail authentication system, a sender, and a receiver to at least partially solve the above-mentioned technical problems.

[0006] To achieve the above objectives, according to a first aspect of this application, a rich detail authentication system is provided, comprising a transmitting end and a receiving end of a communication connection, wherein:

[0007] The receiving end includes: The content processing module is used to perform detail enhancement processing on the video content sent by the sending end in accordance with the determined detail rendering protocol; The effect evaluation module is used to evaluate the detail presentation effect of the video content after detail enhancement processing according to the effect evaluation method in the detail presentation protocol, and generate a detail presentation evaluation result. The credential generation module is used to collect relevant data during the detail enhancement process, combine the detail presentation protocol and the detail presentation evaluation result to generate an authentication credential, and send the authentication credential and the detail presentation evaluation result to the sending end for authentication. The sending end includes: The verification module is used to receive the authentication credentials and detail rendering evaluation results sent by the receiving end after performing detail enhancement on the video content, and to verify the authentication credentials and detail rendering evaluation results based on the detail rendering protocol.

[0008] According to a second aspect of this application, a receiving end is provided, communicatively connected to a sending end. The receiving end includes: The content processing module is used to perform detail enhancement processing on the video content sent by the sending end in accordance with the determined detail rendering protocol; The effect evaluation module is used to evaluate the detail presentation effect of the video content after detail enhancement processing according to the effect evaluation method in the detail presentation protocol, and generate a detail presentation evaluation result. The credential generation module is used to collect relevant data during the detail enhancement process, combine the detail rendering protocol and the detail rendering evaluation result to generate an authentication credential, and send the authentication credential and the detail rendering evaluation result to the sending end so that the sending end can verify the authentication credential and the detail rendering evaluation result based on the detail rendering protocol.

[0009] In one possible design, the receiver further includes: The second session management module is used to negotiate with the sending end to generate a unique session identifier, which is used to associate data transmission and processing operations during the session; The second session management module is also used to incorporate session-related information into the authentication credentials.

[0010] According to a third aspect of this application, a transmitting end is provided, communicatively connected to a receiving end. The transmitting end includes: The verification module is used to receive the authentication credentials and detail rendering evaluation results sent by the receiving end after performing detail enhancement processing on the video content, and to verify the authentication credentials and detail rendering evaluation results based on the detail rendering protocol; wherein, the detail enhancement processing is performed on the video content sent by the sending end according to the established detail rendering protocol, and the detail rendering evaluation results are generated after evaluating the detail rendering effect of the video content after detail enhancement processing according to the effect evaluation method in the detail rendering protocol.

[0011] In one possible design, the transmitter further includes: The first session management module is used to negotiate with the receiving end to generate a unique session identifier, which is used to associate data transmission and processing operations during the session. The first session management module is also used to incorporate session-related information into the authentication credentials and the verification audit log.

[0012] According to a fourth aspect of this application, a rich detail authentication method is provided, applied at a receiving end, the method comprising: Perform detail enhancement processing on the video content sent by the sending end in accordance with the established detail rendering protocol; Based on the effect evaluation method in the detail rendering protocol, evaluate the detail rendering effect of the video content after detail enhancement processing, and generate a detail rendering evaluation result; Collect relevant data during the detail enhancement process, combine the detail rendering protocol and the detail rendering evaluation result to generate an authentication credential, and send the authentication credential and the detail rendering evaluation result to the sending end for authentication.

[0013] According to a fifth aspect of this application, a rich detail authentication method is provided, applied at a receiving end, the method comprising: The receiver receives the authentication credentials and detail rendering evaluation results sent after performing detail enhancement processing on the video content; The authentication credentials and the detail rendering evaluation results are verified based on the detail rendering protocol; wherein, the detail enhancement processing is performed on the video content sent by the sending end according to the established detail rendering protocol, and the detail rendering evaluation results are generated after evaluating the detail rendering effect of the video content after detail enhancement processing according to the effect evaluation method in the detail rendering protocol.

[0014] According to a sixth aspect of this application, an electronic device is provided as a receiver or transmitter, including a memory storing a plurality of instructions; a processor loads instructions from the memory to perform steps in the authentication method provided in the rich detail range of this application.

[0015] According to a seventh aspect of this application, a computer-readable storage medium is provided that stores a plurality of instructions adapted for loading by a processor to perform steps in the authentication method provided in the rich detail range of this application.

[0016] According to an eighth aspect of this application, a computer program product is provided, comprising a computer program or instructions that, when executed by a processor, implement the steps of the authentication method within the rich detail range provided in this application.

[0017] In summary, this application constructs a collaborative authentication system comprising a sender and a receiver. The receiver performs detail enhancement processing on the video content according to a predetermined detail rendering protocol, simultaneously evaluates the processed detail rendering effect and generates an evaluation result. After collecting relevant data from the processing, it combines the protocol and the evaluation result to generate an authentication credential, which is then sent to the sender. The sender verifies the authentication credential and evaluation result based on the detail rendering protocol. This forms a complete technical chain from processing, evaluation, credential generation to verification. This chain establishes a collaborative benchmark between the sender and receiver through the detail rendering protocol, ensuring that the receiver's detail enhancement processing has a clear standard and avoiding the problems of independent enhancement modules and conflicting effects found in existing technologies. Furthermore, the generation and verification mechanism for authentication credentials and evaluation results makes the detail enhancement effect traceable and verifiable, solving the pain point of existing solutions' difficulty in objectively evaluating the effect. Furthermore, direct communication and verification between the sending and receiving ends do not rely on third-party entities. This simplifies the authentication process, achieves precise matching between device capabilities and content requirements, effectively reduces presentation differences between heterogeneous devices, and improves the overall consistency and reliability of rich detail presentation. This provides a systematic solution for high-quality distribution and display of video content.

[0018] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings, wherein the same reference numerals in the following description denote the same parts.

[0021] Figure 1This is a schematic diagram of the structure of an authentication system with a rich range of details provided in an exemplary embodiment of this disclosure; Figure 2 This is a schematic diagram of a scenario of an authentication system with a rich range of details provided in an exemplary embodiment of this disclosure; Figure 3 This is a schematic diagram of the interaction between the sender and receiver in an authentication system provided in an exemplary embodiment of this disclosure. Figure 1 ; Figure 4 This is a schematic diagram of the interaction between the sender and receiver in an authentication system provided in an exemplary embodiment of this disclosure. Figure 2 ; Figure 5 This is a schematic diagram of the interaction between the sender and receiver in an authentication system provided in an exemplary embodiment of this disclosure. Figure 3 ; Figure 6 This is a schematic diagram of the interaction between the sender and receiver in an authentication system provided in an exemplary embodiment of this disclosure. Figure 4 ; Figure 7 This is a schematic diagram of the interaction between the sender and receiver in an authentication system provided in an exemplary embodiment of this disclosure. Figure 5 ; Figure 8 This is a flowchart illustrating a rich detail-oriented authentication method provided in an exemplary embodiment of this disclosure. Detailed Implementation

[0022] In the description of this application, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can mean A or B. "And / or" in this application is merely a description of the relationship between the related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. A and B can be singular or plural.

[0023] In the description of this application, unless otherwise stated, "multiple" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0025] Based on the technical issues mentioned in the background, the relevant background of the rich detail range (RDR) involved in this application will first be introduced.

[0026] The display industry has reached a critical turning point. Over the past decade, the significant improvements in panel resolution, refresh rate, and brightness have far outpaced advancements in content delivery technology. Today's display devices can present a far greater amount of detail than most content in real-world scenarios can provide, especially in streaming and ad-supported distribution environments where resolution and bitrate remain strictly limited. Therefore, image quality is no longer solely constrained by the capabilities of the display device, but also depends on the system's understanding of the displayed content. This application introduces Responsive View Rendering (RDR) as an open, descriptive framework to explore perceived image quality in this new reality, focusing on the richness, integrity, and credibility of visual details. This is particularly relevant when considering AI-driven augmentation processing, examining how these details are reconstructed, preserved, and perceived across spatial, temporal, and structural dimensions. It is important to clarify that RDR does not define standards, certification systems, algorithms, or products, but rather provides a shared language for re-examining and discussing image quality issues as the focus of image quality optimization shifts from signal fidelity to content understanding. This will be elaborated on in the following points.

[0027] First, the shift from signal fidelity to content understanding. Traditional image quality optimization is rooted in signal processing technology. Display devices are designed to faithfully reproduce input pixels as much as possible, primarily improving sharpness, contrast, and noise levels based on the characteristics of the signal itself. This model assumes that image quality begins with the received pixels, but this assumption no longer holds true in modern display systems. Content is often transmitted in compressed, bandwidth-limited formats, and display devices rely on AI-based super-resolution, motion interpolation, and reconstruction techniques to compensate for quality gaps. In this context, the system must determine the objects, locations, and methods of enhancement, decisions that depend on understanding the original content. Today, image quality optimization begins with understanding the content; understanding scene structure, motion dynamics, object boundaries, and perceptual importance is essential. Without this understanding, enhancement algorithms may produce images that are sharp enough but lack stability, rich in detail but lack coherence, or visually striking but perceptually illogical. RDR (Refresh Rate of Rendering) is proposed based on this shift.

[0028] Second, the limitations of existing image quality concepts. Established frameworks such as High Dynamic Range (HDR) have successfully standardized discussions on brightness, contrast, and color space within the industry. These dimensions remain crucial, but they cannot address a growing set of perceptual problems arising from modern enhancement processing workflows. A fully HDR-capable display may still deliver a poor viewing experience in the following situations: fine details fluctuate or "crawl" between frames, reconstructed textures lack structural consistency, motion trajectories appear artificial and unstable, and AI-generated details conflict with scene semantics. These artifacts are difficult to describe using traditional metrics but have a significant impact on perceived image quality. What the industry currently lacks is a way to describe the richness and fidelity of detail in reconstructed scenes—not focusing on the brightness or color of the image, but rather on whether its details are believable, stable, and consistent.

[0029] Third, the definition and boundaries of RDR. First, regarding the definition of RDR, RDR describes the perceived richness and fidelity of visual detail in displayed content across spatial, temporal, and structural dimensions. Its core focus is on how fine details, textures, edges, and motion structures are convincingly preserved or reconstructed, aligned with the scene's geometry, consistent in time, and perceived by a human observer. It is particularly applicable to scenarios where AI-driven processing techniques enhance or reconstruct content.

[0030] Secondly, regarding the boundaries of RDR, to avoid ambiguity, it must be clarified that RDR is not a brightness or contrast indicator, not a replacement for HDR or color standards, not a specific algorithm or processing technology, nor a certification system, rating standard, or product label. RDR is a conceptual and perceptual framework, and its development relies on shared understanding within the industry rather than centralized control.

[0031] Fourth, regarding RDR and HDR (High Dynamic Range).

[0032] RDR and HDR are complementary, not competitive. HDR has successfully standardized discussions on brightness, contrast, and color space within the industry, but HDR alone cannot capture whether visual details (especially those reconstructed by AI) remain perceptually stable and structurally faithful. In practical applications, a display device fully capable of HDR may still exhibit poor perceptual detail quality in the following situations: insufficient spatial resolution of the source content, artifacts introduced by motion reconstruction, and inconsistent or illogical textures produced by AI enhancement. RDR complements HDR by bridging these gaps, focusing on the richness, stability, and credibility of the fine details and motion structures perceived by the viewer. A display device may excel in HDR but still fall short in RDR.

[0033] Fifth, Content Comprehension: The Foundation of RDR In AI-enhanced display processing, enhancement decisions are no longer uniform but depend on the content itself. Effective reconstruction of details requires understanding object boundaries and scene structure, texture and noise characteristics, motion types and trajectories, the importance differences between foreground and background, and the temporal continuity between frames. Without this understanding, enhancement processing becomes guesswork. RDR is based on the core principle that perceiving image quality begins with understanding the content, not just relying on signal processing. RDR focuses not on how much information is transmitted, but on whether the system respects the actual meaning represented by the content.

[0034] Sixth, metadata: a bridge for system collaboration. Once content understanding is achieved, it must be made executable throughout the entire display processing flow. In this context, metadata acts as a bridge, externalizing insights gained from content understanding and enabling coordinated action across different processing stages. Metadata here does not refer to a fixed format or standard, but rather a universal mechanism for conveying guiding information derived from content analysis. By supporting collaboration rather than redundancy, metadata helps ensure that augmented decision-making remains coherent, efficient, and aligned with perception goals. RDR does not specify the representation, encoding, or transmission methods of metadata; multiple implementations can coexist to adapt to different system architectures and constraints.

[0035] Seventh, a new paradigm of image quality In summary, RDR reflects a broader shift in the concept of image quality: understanding content → guiding enhanced decisions → presenting perceptibly credible results. This paradigm recognizes that more data is not always the solution. Under real-world constraints, deeper understanding and more effective collaboration are more important. More detail does not require more data, but rather a more thorough understanding.

[0036] Eighth, an open framework for industry collaboration. RDR was proposed as an open and shared framework; it is not a standard, certification system, or proprietary system. Its purpose is to: facilitate clearer communication among content creators, platform providers, chip suppliers, and display manufacturers; support research related to image quality perception in AI-enhanced scenarios; and encourage diverse implementation methods and continuous optimization. RDR is expected to evolve and develop through industry-wide use and dialogue.

[0037] In view of this, this disclosure provides a rich detail-oriented authentication system, such as Figure 1 As shown, the authentication system includes a sending end and a receiving end with a communication connection. The sending end can provide video content to the receiving end (see...). Figure 2 ).

[0038] Among them, such as Figure 3 As shown, the receiving end includes a content processing module, an effect evaluation module, and a certificate generation module. The content processing module is used to perform detail enhancement processing on the video content sent by the sending end according to the established detail rendering protocol.

[0039] The aforementioned effect evaluation module is used to evaluate the detail rendering effect of the video content after detail enhancement processing according to the effect evaluation method in the detail rendering protocol, and generate detail rendering evaluation results.

[0040] The aforementioned credential generation module is used to collect relevant data during the detail enhancement process, combine the detail rendering protocol and detail rendering evaluation results to generate authentication credentials, and send the authentication credentials and detail rendering evaluation results to the sending end for authentication.

[0041] The sending end includes a verification module, which is used to receive the authentication credentials and detail presentation evaluation results sent by the receiving end after enhancing the details of the video content.

[0042] For example, the receiving end described above is a device that displays video content. The receiving end can be a smart TV, set-top box, streaming media player, XR / AR headset, car display, game console, or any device that performs real-time display optimization. In other embodiments, the receiving end can be an entire device, a system-on-a-chip (SoC) within the device, or a timing controller (TCON).

[0043] The sending end can be a device providing video content, a device providing video services, a cloud server, an edge computing node, or an encoder. The sending end can be a server (such as a cloud server on a video platform), a content owner, an edge computing node, etc.

[0044] In addition, the certification system may include a platform or orchestrator for coordinating negotiation and governance across heterogeneous device categories, managing policy levels, standardizing report formats, and aggregating compliance statistics.

[0045] It should be noted that the authentication system can be deployed in a fully distributed, centralized, device-based, or separate manner. A fully distributed deployment means the sender and receiver interact directly without the need for other devices such as platforms or orchestrators. A centralized deployment is driven by a platform or orchestrator, which controls the negotiation process and policy management. A device-based deployment allows the receiver to independently complete local scoring and closed-loop optimization without additional support from the sender. A separate deployment provides optimization suggestions and intent guidance, and the receiver adapts and adjusts autonomously based on its own capabilities.

[0046] In this embodiment, the content processing module at the receiving end can first obtain the Detailed Rendering Protocol (RDR contract) negotiated and agreed upon with the sending end. This RDR contract includes a set of allowed processing operations. The content processing module can then invoke the Real-Time Enhancement Pipeline to perform detail enhancement processing on the decoded video content transmitted by the sending end, according to the RDR protocol.

[0047] Subsequently, the effect evaluation module, in accordance with the effect evaluation method (i.e. RDR measurement conditions) agreed in the detail rendering protocol, calls the RDR scoring engine to evaluate the enhanced video content and obtain the detail rendering evaluation result (i.e. RDR score) to quantitatively assess whether the enhancement effect is compliant.

[0048] The credential generation module collects relevant data from the detail enhancement process (such as the model version used for enhancement, parameter configuration records, and processing resource consumption (such as computational budget usage and latency data)), combines it with the detail presentation protocol and the aforementioned detail presentation evaluation results, and packages it to generate authentication credentials (i.e., the RDR evidence package). Afterward, the credential generation module sends the authentication credentials and the detail presentation evaluation results to the sending end.

[0049] After receiving the authentication credentials and presentation details assessment results, the sending end's verification module verifies the authentication credentials and presentation details based on the provisions of the presentation details protocol negotiated and agreed upon with the receiving end. It assesses the integrity of the credentials and whether the presentation details meet the standards, ultimately providing a clear judgment of compliance or non-compliance. This forms a complete technical chain from processing, assessment, credential generation to verification.

[0050] In this embodiment, the content processing module can extract unprocessed basic video data from the video stream corresponding to the received video content through the decoder and metadata parser at the receiving end. Under hardware resource constraints, the real-time enhancement pipeline at the receiving end executes multiple enhancement stages in a preset order, such as super-resolution, denoising, deblocking, sharpening / detail reconstruction, tone mapping, and motion-related processing stages. Among them, the super-resolution processing stage amplifies the video resolution and supplements lost detail information through algorithms; the denoising processing stage suppresses random noise and interference signals in the video; the deblocking processing stage eliminates blocky distortion caused by video compression; the sharpening / detail reconstruction stage enhances the clarity of video edges and textures; the tone mapping stage adjusts the color gamut and contrast of the video to adapt to the characteristics of the display panel; and the motion-related processing stage optimizes the continuity of details during object movement. The processing parameters of each stage can be dynamically adapted according to the characteristics of the video content.

[0051] Additionally, the receiving end can include an ROI (Region of Interest) adaptive execution module. This module prioritizes important regions in the video content (such as faces, text, and foreground objects). After identifying high-priority regions using semantic segmentation algorithms, it assigns them computationally more complex processing models and optimized parameter configurations. Simultaneously, a protective band is set at the boundary between high-priority regions and other regions, and a hybrid transition algorithm is used to avoid obvious boundary differences. In the temporal dimension, inter-frame smoothing reduces fluctuations in details within high-priority regions, preventing flickering and ensuring overall image stability.

[0052] Furthermore, the effect evaluation module can collect enhanced video data in real time based on local scoring and closed-loop control, following the effect evaluation methods agreed upon in the detail rendering protocol. It extracts detail rendering-related features such as edge sharpness and texture integrity, and combines this with the detection results of poor rendering effects to generate a detail rendering evaluation result. If the evaluation result does not meet the compliance requirements stipulated in the protocol, the effect evaluation module can analyze the specific reasons for the failure. The optimization and adjustment module then adaptively adjusts the algorithm selection, parameter configuration, or resource allocation strategy for enhancement processing, re-executing the enhancement processing flow until the evaluation result meets the requirements or reaches the preset iteration limit. The entire process is conducted under the constraints of frame time budget and segment time budget to avoid affecting the smoothness of video playback.

[0053] In some embodiments, the aforementioned RDR contract may include multiple of the following: core identifier, compliance objectives and thresholds, measurement conditions and rules, hardware constraints, permitted sets of operations and policies, rollback and renegotiation rules, reporting and evidence requirements, and security and integrity related information.

[0054] The aforementioned core identifiers include at least one of the following: contract ID, schema version (structural definition version, such as 1.0), effective time range, and associated identifier.

[0055] The aforementioned contract ID and model version can be understood as the contract number plus the version number, to confirm that the sender and receiver are using the same RDR contract. The aforementioned effective time range indicates the scenario and time period in which the RDR contract is effective. The effective time range can include at least one of the applicable periods for sessions, segments, and content. A session refers to a conversation between the sender and receiver. Content refers to the video (such as a movie) transmitted from the sender to the receiver, which is the object of detail enhancement processing. A segment is a portion of the content, such as a continuous sub-fragments of content split along a time dimension. The aforementioned associated identifiers can include at least one of the following: session ID, evidence ID, text ID, segment ID, device ID, device type ID, pipeline ID, and model setting ID.

[0056] Session ID: A session identifier between the sender and receiver, which may include identifiers for playback or negotiation sessions. Text ID, also known as Content ID (content_id), represents the content title or asset identifier. Evidence ID: An identifier for an evidence package or RDR certificate. Segment ID (segment_id): An identifier for a time segment, scene, group of pictures (GOP), or data block. Device ID, Device Category ID: A stable device identifier or category identifier (which may use privacy-preserving variants). Optionally, for privacy protection, the device ID may not use a hardware address, but instead use a device identifier based on device category or a rotation-based session identifier. Pipeline ID: An identifier for a series of detailed enhancement processing flow configurations. Model Setup ID: An identifier for a set of available or used model versions. It is understood that the above core identifiers are generic, as they are common in messages and logs.

[0057] The aforementioned compliance goals and thresholds may include at least one of the following: overall goals, sub-score constraints, and compliance rules. Overall Goals: These are the quantitative standards for detail enhancement compliance, i.e., the level of detail presentation. They may include a target score or target score band (e.g., [80, 95]) or grade (high, medium, low), and a minimum acceptable score (or minimum score). Sub-score Constraints: These quantify the level of detail presentation from different dimensions, including at least one of the following: minimum threshold for spatial detail, minimum threshold for temporal detail, and maximum threshold for artifact penalty. Compliance Rules: These define the standards for meeting detail presentation requirements, such as 90% or more of the segment scores being greater than or equal to 80, and the worst segment score being no less than 75. The aforementioned measurement conditions and rules may include at least one of the following: measurement mode, sampling rules, aggregation rules, ROI weighted rules, and scoring versions. Measurement Modes: These include at least one of the following: frame, segment, window, ROI, and tile. For example, measurement every 3 minutes. Sampling Rules: These include at least one of the following sampling methods: uniform sampling, content-adaptive sampling, ROI-prioritized sampling, and random audit sampling. Sampling-based evaluation can avoid excessive resource consumption from frame-by-frame measurement. Uniform sampling involves periodically acquiring frames and blocks. Content-adaptive sampling involves obtaining more samples in complex or high-motion segments.

[0058] ROI Prioritized Sampling: Higher sample density in critical areas. Random Audit Sampling: Randomly sampling frames and blocks, i.e., unpredictable sampling points, to prevent gameplay. Aggregation Rules: The method of representing statistical scores, including score aggregation methods (average, percentiles, worst-case), aggregation windows (N frames, specific duration), etc. For example, the segment score is the average of all sampled frames within that segment. ROI Weighting Rules: Includes ROI category definitions (e.g., face, text, foreground, background) and the weight allocation for each ROI. Scoring Version: Specifies the applicable RDR scoring algorithm version and calibration parameters. The above hardware constraints can include at least one of performance constraints, resource constraints, and effect constraints. Performance Constraints: Can include latency caps (per frame, per segment deadline), computation budget (NPU / GPU cycles / normalized computation tokens), etc.

[0059] Resource constraints: These may include at least one of the following: power consumption threshold, memory bandwidth limit, and network bandwidth limit (including metadata budget). Effect constraints: These may include hard limits on artifacts (such as the maximum allowed penalty value for each artifact type, including ringing, halos, flickering, and striping). The allowed set of operations and strategies mentioned above may include at least one of the following: module permissions, model restrictions, parameter boundaries, ROI strategies, and scheduling constraints. Module permissions: Allowed detail enhancement modules (such as superresolution (SR), noise reduction (NR), unblocking, sharpening, motion estimation and motion compensation (MEMC), and tone mapping), i.e., the allowed set of operations. Model restrictions: Allowed model families and model version ranges.

[0060] Parameter boundaries: The value range of adjustable parameters for each module (e.g., sharpening intensity, denoising threshold, time smoothing window). ROI strategy: ROI layering rules, tile size, overlap range, seam suppression requirements, etc. Scheduling constraints: Module processing order, parameter adjustment step size limits, and model switching frequency limits to avoid oscillations.

[0061] The aforementioned rollback and renegotiation rules may include at least one of the following: rollback tiers, triggering conditions (i.e., protocol update triggering conditions), and execution rules. Rollback Tier: The order of target degradation when constraints tighten (e.g., high → medium → low), and the upper limit of artifacts to be retained during rollback. For example, in the aforementioned hot throttling, if the receiver temperature is >60℃, the overall target is reduced from a high level to a low level, but the upper limit of artifacts remains ≤10. Triggering Conditions: Clearly define the triggering conditions for rollback and renegotiation, including at least one of hot throttling, power mode change, bandwidth decrease, user mode switching, content type conversion, and continuous compliance failure. Execution Rules: The timing of renegotiation (immediate / segment boundary / scene cut-off), and the grace period (the allowed duration / number of frames before renegotiation).

[0062] The aforementioned reporting and evidence requirements are used to instruct the recording and sharing of evidence, which may include multiple of the following: obtained scores (such as the RDR score mentioned above), sub-scores, confidence levels, compliance results, contract identifiers, measurement conditions, equipment IDs, equipment category IDs (or equipment class IDs), pipelines, model versions, score tracking, and action tracking.

[0063] For example, reporting and evidence requirements may include at least one of the following: reporting level, evidence fields, retention rules, and audit support. Reporting level: The granularity at which reporting is required, such as segmented compliance results only, frame-level segmentation, or ROI-level segmentation. Evidence fields: Information that must be recorded, such as score trajectories, action history, constraint status, sampling records, and pipeline fingerprints. Retention rules: Evidence storage duration and shareable scope (local storage or upstream reporting). Audit support: Whether offline auditing is supported, and the method of providing evidence packages (on-demand transmission or real-time reporting).

[0064] The aforementioned security and integrity-related information may include at least one of integrity protection, non-repudiation, and version binding. Integrity protection: the hash and signature information of the contract itself. Non-repudiation: signature requirements and timestamp specifications for the evidence log. Version binding: the bound scoring engine version and optimizer version verification rules.

[0065] In some embodiments, the above-described detail enhancement processing includes at least one of spatial detail enhancement operation, temporal detail stabilization operation, and poor rendering effect suppression operation.

[0066] Spatial detail enhancement operations include at least one of multi-scale detail reconstruction, edge enhancement, and texture restoration. Multi-scale detail reconstruction is used to restore content details at different scales, edge enhancement is used to improve the clarity of content edges, and texture restoration is used to improve the texture information in the content.

[0067] Temporal detail stabilization operations include at least one of motion consistency optimization, inter-frame detail alignment, and flicker suppression. Motion consistency optimization ensures the continuity of detail during the movement of displayed objects, inter-frame detail alignment eliminates detail discrepancies between frames, and flicker suppression prevents periodic brightness or detail fluctuations in the content.

[0068] Undesirable rendering suppression operations include at least one of ringing suppression, halo removal, banding removal, and noise suppression. Ringing suppression is used to eliminate ringing artifacts near edges, halo removal is used to remove halo effects caused by over-enhancement, banding removal is used to eliminate color banding or brightness banding in the content, and noise suppression is used to reduce random noise in the content.

[0069] In this embodiment, the detail enhancement processing is performed by a detail enhancement module. Different detail enhancement modules can perform different detail enhancement operations. Exemplary non-limiting control measures (i.e., detail enhancement operations) performed by the super-resolution / detail reconstruction module may include: model series selection (light / medium / heavy), magnification and refinement intensity, iterations / steps (for iterative models), precision mode (e.g., reduced speed precision), ROI-aware invocation (applying heavier SR only to high-priority regions), and artifact suppression mode (e.g., detail-first, artifact-safe). When the spatial detail score is low, the RDR optimizer may increase the SR intensity but respect the artifact cap (ringing / halo / illusion penalty).

[0070] The denoising / deblocking module can implement unrestricted control measures, including: denoising intensity (global and specific ROI), deblocking intensity (content-adaptive), detail preservation mode (edge-aware denoising), and temporal denoising settings. ROI refers to achieving high quality in key areas (such as ROI areas) through differentiated resource allocation, avoiding wasted computing power in irrelevant areas (such as solid color backgrounds).

[0071] When artifact penalties indicate that compression noise is amplified, the optimizer may increase denoising / deblocking while ensuring that it does not excessively suppress true details.

[0072] Non-restrictive controls performed by the sharpening / local contrast enhancement module may include: enhancement strength and radius, edge threshold and halo suppression threshold, local contrast gain limitation, and ROI gating (e.g., avoiding sharpening skin tones).

[0073] When ringing / halo penalties increase, the optimizer can reduce sharpening, or selectively apply sharpening only if the gains in spatial detail outweigh the risk of artifacts.

[0074] The non-restrictive controls performed by the tone mapping / HDR processing module may include: tone curve selection, local tone mapping intensity and smoothing, banding suppression mode, and highlight / shadow detail preservation settings. By adjusting the tone mapping, perceived detail in highlights and shadows is preserved while avoiding banding penalties.

[0075] Non-restrictive controls performed by the motion processing / temporal stabilization / MEMC module may include: motion estimation quality level, motion compensation refinement intensity, temporal smoothing intensity and window size, flicker suppression threshold, frame interpolation enable / disable, and ROI / motion gating (e.g., stronger temporal regularization in high motion ROIs).

[0076] When timing details become unstable or flicker penalties increase, the optimizer can add timing consistency control.

[0077] Non-restrictive control measures performed by the device (such as panel, TCON) perception optimization module may include: panel-specific sharpening or subpixel adjustments, overspeed / response time compensation limits, local dimming interactive settings, anti-contour / banding control, and spatiotemporal shapes that match the panel's response characteristics.

[0078] In some embodiments, the detail rendering effect evaluation process performed by the above-described effect evaluation module may include: acquiring image data of the video content after detail enhancement processing according to the evaluation method in the detail rendering protocol; extracting detail rendering-related features from the image data, including at least one of edge sharpness features, texture distribution features, color accuracy features, and motion consistency features; quantitatively evaluating the detail rendering-related features based on the detail rendering standards in the detail rendering protocol to obtain a detail rendering effect index; detecting poor rendering effects in the image data, identifying and quantifying the type and degree of the detected poor rendering effects to obtain a poor rendering effect index.

[0079] By combining the detail rendering effect index and the poor rendering effect index, a detail rendering evaluation result is generated. The detail rendering evaluation result includes the detail rendering status of each display area and the distribution of poor rendering effects.

[0080] For example, the receiving end determines the acquisition range based on the measurement mode in the RDR contract and acquires image data based on the sampling rules in the RDR contract. For instance, if the RDR contract stipulates that the video is segmented every 3 minutes with ROI priority sampling, then the video is divided into 3-minute time units, and image frames of high-priority ROI areas such as faces, text, and foreground are acquired first, while background frames are acquired according to uniform sampling rules, such as acquiring 1 background frame every 10 frames, to ensure coverage of the entire display area.

[0081] In addition, each frame of image data is bound to a timestamp, region identifier (such as ROI category, background, etc.), and sampling rules, and the bound data is stored as structured data for subsequent partition evaluation and auditing.

[0082] After acquiring image data of the video content, the effect evaluation module performs feature extraction on the acquired image data (or the structured data mentioned above) to present relevant features covering at least one of the details of edge sharpness, texture integrity, color accuracy, and motion consistency.

[0083] The effectiveness evaluation module quantifies the extracted detail presentation-related features based on the detail presentation standards (i.e., RDR scoring standards, which are also compliance targets and thresholds) in the detail presentation protocol, and obtains the detail presentation effectiveness index (i.e., RDR score).

[0084] Next, the effect evaluation module uses a built-in artifact detector, or artifact detection algorithm, to detect undesirable rendering effects in the image data, and identifies and quantifies their type and degree to obtain an undesirable rendering effect index. The detection of undesirable rendering effect types can include: identifying ringing artifacts (ghosting near edges), halo effects (blurred halos at edges caused by over-enhancement), banding (color banding, brightness banding, etc.), noise (random noise, mosquito noise, etc.), flicker (periodic brightness, detail fluctuations, etc.), and phantom textures (synthesized fake textures), among other undesirable types constrained by the protocol.

[0085] Quantification of the degree of defect presentation: For each type of defect detected, the degree of defect is quantified according to the penalty standards specified in the RDR contract. For example, ringing artifacts are divided into 5 levels according to the width and intensity of the ghosting, with level 1 being the lightest and level 5 being the most severe. Different levels correspond to different penalty scores, such as level 1 penalizing 5 points and level 5 penalizing 25 points.

[0086] Generate defect rendering index: Calculate the quantitative penalty score of all defect types to obtain the global and individual display area defect rendering index (i.e., artifact penalty P_artifact), and record the distribution location of defect rendering effects, such as ringing at the edge of a certain ROI or stripes in the background area.

[0087] Then, the performance evaluation module can calculate the RDR total score according to the formula RDR = w_s S_spatial + w_t S_temporal - w_a P_artifact, where w_s, w_t, and w_a are the weights agreed upon in the protocol, is used to calculate the total score for the global detail rendering effect. S_spatial represents the spatial detail score corresponding to the spatial detail-related features, and S_temporal represents the temporal detail score corresponding to the motion consistency feature score.

[0088] It also categorizes regions by ROI and background area, outputting sub-scores for the detail rendering effect of each region (e.g., facial region edge clarity 88 points, texture integrity 85 points) and the distribution of poor rendering effects (type, degree, location coordinates).

[0089] As an example, the process of quantifying the relevant features of the above-mentioned details can be performed by an RDR scoring engine (or scorer). The role of the RDR scoring engine is to quantify the richness and stability of fine details after video processing under explicit measurement conditions, while penalizing bad artifacts. Its output is both operable (for subsequent closed-loop optimization), comparable (i.e., consistent across devices), and auditable (supported by evidence packages).

[0090] The score generated by the RDR scoring engine reflects whether the image data, after detail enhancement processing, maintains a rich detail experience while keeping artifacts within an acceptable range. This score can be evaluated for each frame, each region of interest, each tile, each segment, or each session. Furthermore, this evaluation process can be performed at the receiver or transmitter, or in a separate configuration, supporting measurement modes with no reference, partial reference, or auxiliary information assistance.

[0091] The overall RDR score is calculated through a weighted combination of sub-scores and can include three parts: spatial detail score, temporal detail score, and artifact penalty. The spatial detail score measures the richness and preservation of multi-scale details (such as edges, textures, and microstructures). The temporal detail score measures the stability and temporal coherence of motion-consistent details. The artifact penalty deducts undesirable artifacts such as halos, ringing, flickering, banding, or inconsistencies in composite detail; its calculation formula can be expressed as RDR = w_s. S_spatial + w_t S_temporal - w_a P_artifact. The weights can be fixed by the rating version, configured by contract, or selected based on content category.

[0092] In addition to a single scalar score, the RDR scoring engine can also output a scoring vector (e.g., {S_spatial, S_temporal, P_artifact, confidence}). In this case, compliance must meet multiple constraints, including spatial detail not falling below a threshold (S_spatial ≥ T_spatial), temporal detail not falling below a threshold (S_temporal ≥ T_temporal), artifact penalty not exceeding a threshold (P_artifact ≤ T_artifact), and overall score not falling below a threshold (overall score ≥ T_overall). This vectorized form effectively ensures no artifact amplification while maximizing detail. Here, confidence is a reliability metric output by the RDR scoring engine, reflecting the credibility of the current RDR score and evaluation results.

[0093] As an example, the aforementioned spatial detail score primarily evaluates the presentation and credibility of fine structures. Its implementation methods can include at least one of multi-scale structural persistence, region-aware detail assessment, structural credibility constraints, and compression rebound metrics. Multi-scale structural persistence evaluates edge and texture energy across scales using methods such as pyramid or wavelet-like analysis, rewarding consistent and non-fake details across different scales. Region-aware detail assessment applies different weights to different levels of interest regions, such as faces, text, foreground objects, and backgrounds, favoring detail improvement in areas that are visually sensitive to humans or have clear content intent. Structural credibility constraints penalize fictitious or inaccurate textures that do not conform to structural cues, such as unnatural repeating patterns or inconsistent edges. The compression rebound metric distinguishes between recovered details and amplified compressed noise, avoiding spurious detail gains. Furthermore, the spatial detail score can be determined solely based on detail presentation-related features, or it can be calculated by combining other content, such as content side information in metadata.

[0094] Temporal detail scoring measures the stability of detail over time, avoiding the introduction of flickering or motion-inconsistent textures. It can include at least one of motion-consistent detail tracking, temporal stability and anti-flicker measures, temporal boundary consistency, and motion-sensitive weighting. Motion-consistent detail tracking tracks fine structures along the estimated motion path and rewards consistency. Temporal stability and anti-flicker measures penalize inter-frame oscillations in sharpness or texture intensity in scenes such as static areas. Temporal boundary consistency penalizes visible seams that change over time due to tile-level model switching or boundary blending instability. Motion-sensitive weighting imposes a higher penalty for artifacts in high-motion or high-attention-salience regions.

[0095] Additionally, time scoring typically operates over a sliding window (such as N frames or a specific duration) and outputs the mean, variance, or worst-case indicator.

[0096] The RDR scoring engine also includes one or more artifact detectors used to generate artifact penalties. Furthermore, the artifact penalties are combined with upper limits defined in the RDR contract; exceeding any of these upper limits, regardless of spatial or temporal detail gains, is considered non-compliant.

[0097] The RDR scoring engine can operate under different reference regimes, including at least one of no-reference mode, partial-reference mode, and auxiliary information mode. No-reference mode uses only image data with detail enhancement and an artifact detector, suitable for real-time scoring. Partial-reference mode uses sparse references, such as reference patches, anchor frames, or content-side samples, balancing accuracy and bandwidth. Auxiliary information mode utilizes metadata to improve scoring reliability, distinguishing true details from noise or artifacts. The scoring version explicitly indicates the currently active reference regime and the output interpretation method.

[0098] Scores can be statistically analyzed and aggregated across multiple dimensions, such as using averages, percentiles, and worst-case scenarios to calculate scores from frame to segment, from ROI to ROI-weighted scores, and from segment to session scores (i.e., compliance rate). Aggregation rules can be fixed based on the scoring version or specified in the RDR contract. Furthermore, the RDR scoring engine outputs the aforementioned confidence level, which can be determined based on dimensions such as sample count and coverage, content type stability, differences between different frames or different ROIs, and known limitations of the selected reference regime. For example, the confidence level can be used to trigger additional sampling, renegotiation, or conservative fallback strategies.

[0099] Optionally, to reduce sensitivity to game behavior, i.e. to avoid compromising perceived quality while optimizing scores, the RDR scoring engine also includes several robustness mechanisms, such as multi-component constraints to limit detail enhancements to not exceed artifact limits, random or partially hidden audit sampling rules, multiple scoring versions or calibration sets across content categories, compliance requirements based on percentiles or worst-case scenarios (rather than just averages), and cross-checks to distinguish between true detail restoration and noise amplification.

[0100] At the output level, the RDR scoring engine's output is divided into two layers: control output and authentication output. The control output is directed to the optimizer and includes the score vector, gradient, proxy, and feedback for each component. The control layer output explicitly indicates what needs to be adjusted, such as reducing ringing or increasing texture.

[0101] The certification output is geared towards evidence generation, including segmented compliance metrics, measurement context, sampling records, version identifiers, and trace summaries that can be used to generate RDR certificates / evidence packages.

[0102] The RDR scoring engine allows for flexible adjustments to execution steps based on actual conditions such as device capabilities and scenario requirements (e.g., network, power consumption), while adhering to the RDR contract, and does not restrict other compliance implementation methods. For example, implementation methods may include real-time scoring, segmented scoring, and platform-audited scoring. In real-time scoring, the receiving end calculates the no-reference score for each segment, adjusts parameters, and records an evidence summary. In segmented scoring, the sending end calculates a partial reference anchor score, and the receiving end calculates a no-reference operation score; both are combined in the evidence package. In platform-audited scoring, the platform requires evidence and recalculates the verification score of a subset of samples to verify the certificate.

[0103] Taking the aforementioned details, including edge sharpness, texture distribution, color accuracy, and motion consistency, as an example, the process of extracting these details can include: Edge detection algorithms are used to extract edge information of displayed objects from image data, generating edge sharpness features. Texture analysis algorithms are used to extract texture distribution features from image data, including at least one of texture density, texture direction, and texture repetition pattern. Through color contrast analysis, the image color information in the video content after detail enhancement is compared with the image color information in the original video content before detail enhancement, calculating the degree of color deviation and generating color accuracy features. Based on inter-frame motion analysis, the detail changes of displayed objects in consecutive frames are tracked, and the consistency of details during motion is evaluated, generating motion consistency features. Edge sharpness features, texture distribution features, color accuracy features, and motion consistency features are standardized to use a unified quantization dimension to generate detail presentation evaluation results.

[0104] In this embodiment, the effect evaluation module can employ an edge detection algorithm to accurately extract edge information of displayed objects in the image data. Through multi-scale processing methods such as pyramid decomposition or wavelet-like analysis, the gradient intensity, continuity, and sharpness of edges at different scales are calculated, and high-contrast edges, such as text edges and object outlines, are selected. This generates edge sharpness features that can quantify edge discrimination capabilities, ensuring that the edge sharpness features reflect the edge rendering quality at multiple scales.

[0105] The performance evaluation module uses a texture analysis algorithm to extract texture distribution features from image data. During the extraction process, based on structural credibility constraints, it distinguishes between real textures and amplified compressed noise through texture energy calculation, local binary pattern analysis, and other methods, eliminating false information and ensuring that the texture distribution features contain only credible detail information, thereby accurately reflecting the richness and consistency of image texture.

[0106] The effect evaluation module generates color accuracy features through color contrast analysis. It compares the color information of the enhanced image with the color information of the original video content before processing, channel by channel (such as R channel, G channel, and B channel). It calculates the offset of color channels, white balance deviation, and color gamut coverage difference, and quantifies the degree of color deviation. This is to avoid the color deviating from the original content intention due to over-enhancement, and to ensure that the color accuracy features can effectively reflect the impact of enhancement processing on color fidelity.

[0107] In terms of motion consistency feature extraction, the performance evaluation module is based on inter-frame motion analysis technology and a detail tracking mechanism. It performs motion estimation on continuously acquired frame sequences to obtain the motion path of the displayed object. By tracking the continuity of details on the motion path and calculating the inter-frame detail alignment error, it quantifies whether details exhibit jitter, blurring, or inter-frame inconsistency during the motion process. This generates motion consistency features that reflect the stability of details over time and ensures consistency at temporal boundaries.

[0108] Finally, the effect evaluation module standardizes edge sharpness features, texture distribution features, color accuracy features, and motion consistency features. For example, based on score vector normalization rules, the effect evaluation module maps the quantification results of various features to a unified quantification dimension, eliminating dimensional differences between different features. This ensures that all features are fairly weighted according to the weights agreed upon in the RDR contract when generating subsequent detail rendering evaluation results, providing a unified standard for accurately quantifying detail rendering effects.

[0109] In some embodiments, the aforementioned RDR contract, also known as the Presentation Details Protocol, is negotiated and determined between the sender and receiver. See also Figure 4 The sending end also includes an intent generation module and a negotiation initiation module, while the receiving end may include a negotiation response module.

[0110] The intent generation module is used to generate detailed presentation intents, which include detailed presentation standards and limitations on poor presentation effects for video content.

[0111] The negotiation initiation module is used to negotiate capabilities with the receiving end based on the detailed presentation intent and determine the detailed presentation protocol.

[0112] The negotiation response module is used to respond to the sender's negotiation request based on the receiver's capability description information, and to jointly negotiate and determine the detailed presentation protocol with the sender. The capability description information includes supported content processing methods and device operating constraints.

[0113] As an example, Detail Representation Intent (or RDR Intent) is the desired target generated by the sender for video content, including the detail rendering standard and restrictions on undesirable rendering effects for that video content. For example, for movie-type video content, the detail rendering standard can be set to a spatial detail score of no less than 85 points and a temporal detail score of no less than 80 points, while the restrictions on undesirable rendering effects can be set to a ringing artifact penalty of no more than 10 points and a flicker artifact penalty of no more than 5 points. As another example, for sports event video content, the threshold for temporal detail score can be increased to 85 points, while the upper limit for artifacts in motion scenes can be lowered.

[0114] For example, the above-mentioned intent to present details may include at least one of the following information: Core Objective Setting: Includes at least one of the following: target score, target score band, and minimum acceptable score. Detail vs. Artifact Trade-off Preference: Clearly defines the preference between detail enhancement and undesirable artifacts, with options for detail priority, balanced mode, and artifact safety. For example, artifact safety can be chosen for text content, while detail priority can be chosen for sports events. ROI-Related Rules: Defines the ROI tiers, ROI weight allocation, and priority rules, such as prioritizing detail quality for high-level ROIs. Artifact Cap Constraint: Sets the maximum allowed component (i.e., upper limit) of undesirable artifacts. Latency Budget: The maximum latency limit for detail enhancement processing to ensure a real-time experience. Allowed Operation Set: Limits the detail enhancement operations that the receiver can perform, such as only super-resolution (SR), SR, and denoising. Reporting Level Requirements: Clearly defines the granularity of the evaluation results that the receiver needs to report. Privacy Rules: Specifies the range of device identifiers and log data that can be retained or shared, such as only allowing the sharing of device class IDs and prohibiting the transmission of unique device identifiers. Optimize auxiliary seeds: including suggested initial strategy ID, ordered list of candidate models (such as lightweight SR model and medium SR model), parameter adjustment limits (such as sharpening intensity 0.3-0.8), denoising intensity, ROI enhancement hints (such as focusing on enhancing text region texture), and ordered fallback steps.

[0115] As an example, the capability description information above represents the capabilities of the receiving end related to video processing and display, which may include supported content processing methods and device operating constraints.

[0116] Supported content processing methods refer to the set of technical means available to the receiving end for performing video detail enhancement, covering modules, models, parameter configurations, and regional scheduling strategies, with the goal of achieving spatial detail enhancement, temporal detail stabilization, and suppression of unwanted artifacts.

[0117] Device operating constraints refer to the hardware, software, and environmental limitations that the receiving end must adhere to when performing content processing, determining the actual feasible scope of the content processing method. Simply put, capability description information describes what operations the receiving end can perform and what operations it can perform under certain constraints.

[0118] Optionally, the capability description information includes static capability information and dynamic constraint information of the receiver. The static capability information includes the resolution supported by the receiver, refresh rate, the type of processing module used to perform detail enhancement processing (i.e., the detail enhancement processing module type), and model version. The processing module type includes at least one of a super-resolution processing module, a denoising processing module, a deblocking processing module, a sharpening processing module, a tone mapping processing module, and a motion processing module.

[0119] Dynamic constraint information includes at least one of the following: the receiver's current thermal state, power mode, computational margin, and memory bandwidth.

[0120] Correspondingly, the negotiation response module is also used to integrate static capability information and dynamic constraint information to form capability description information and send it to the sending end for capability negotiation with the sending end.

[0121] Optionally, static capability information refers to the inherent and long-term unchanging attributes of the receiver, including supported resolutions, refresh rates, supported detail enhancement processing module types, model versions, and basic hardware parameters.

[0122] Dynamic constraint information: Real-time changing operating status, including current thermal state, power mode, instantaneous computing margin, memory bandwidth utilization, network transmission status, etc.

[0123] Specifically, capability description information may include at least one of the following: Basic device and platform description: including device category ID, hardware revision, operating system version, and supported enhancement pipeline family identifiers (i.e., detailed enhancement processing module identifiers).

[0124] Display and panel constraints: maximum resolution, maximum refresh rate, supported HDR modes (such as HDR10, Dolby Vision), and panel constraints (such as peak brightness limit, local dimming mode support).

[0125] Computation and real-time constraints: include frame time budget (e.g., processing ≤8ms per frame), computation budget (abstract unit or number of normalized tokens), power mode options (e.g., normal, eco, turbo), and current thermal state (e.g., whether it is in throttling state).

[0126] Algorithm and model support: Explicitly supports the types of detail enhancement modules, model set IDs and model versions (e.g., SR model V3.0), and precision modes (e.g., FP16, INT8).

[0127] I / O and bandwidth capabilities: Mark memory bandwidth levels and network status snapshots (such as current transmission rate and latency) to ensure compatibility between data processing and transmission.

[0128] Measurement-related support: This section describes the supported scoring versions (e.g., scoring algorithm V1.2) and measurement mode options (e.g., frame-level, segment-level, ROI-level sampling, uniform sampling, adaptive sampling, and random audit sampling).

[0129] As an example, the sender can also send metadata about the video content to assist the receiver in performing detail enhancement processing. Correspondingly, such as... Figure 5As shown, the sending end also includes a metadata generation module, used to analyze the video content, extract scene structure cues from the video content, and generate metadata to instruct the sending end to perform detail enhancement processing on the video content. The metadata includes at least one of the following: edge cues, motion cues, depth cues, and important region identifiers. The negotiation initiation module of the sending end is also used to combine the metadata and detail presentation intentions... Figure 1 It is then sent to the receiving end for negotiation and confirmation of the detailed presentation protocol.

[0130] In this embodiment, the metadata generation module at the sending end performs in-depth analysis of the video content, extracts scene structure clues, and generates metadata to instruct the receiving end to perform detail enhancement processing. This metadata includes at least one of edge clues, motion clues, depth clues, and important region markers. Edge clues can annotate the positional information of the outlines of people and the boundaries of objects in the video frame; motion clues record the motion trajectory and speed of fast-moving objects in the picture; and important region markers accurately mark visually sensitive areas such as faces and text. This metadata can provide precise guidance for the detail enhancement processing at the receiving end, thereby improving the presentation effect of the detail enhancement processing at the receiving end.

[0131] Optionally, the negotiation initiation module is also used to send a negotiation request to the receiving end through the negotiation channel. The negotiation request includes detailed presentation intent and metadata.

[0132] The negotiation response module is used to respond to negotiation requests and feed back capability description information to the negotiation initiation module.

[0133] The negotiation initiation module is also used to analyze the degree of matching between the intent to present details and the capability description information, determine the level of detail presentation that the receiving end can achieve under the current capabilities, exchange opinions with the receiving end, determine the compliance requirements, effect evaluation methods and adjustment rules of the detail presentation protocol, generate the detail presentation protocol, and synchronize the detail presentation protocol to the receiving end.

[0134] As an example, the negotiation channel, also known as an in-band signaling channel, side channel, or control plane message channel, is a communication channel between the sender and receiver used to transmit negotiation-related data, such as rendering intent, metadata, capability description information, and rendering protocol details.

[0135] Compliance requirements include the level of detail in the presentation of objectives, the scope of prohibited adverse presentation effects, and rules for allocating processing resources. Effectiveness evaluation methods include the scope of evaluation, evaluation frequency, and reference benchmarks. Adjustment rules include strategies for adjusting processing methods and triggering conditions for protocol updates.

[0136] In this application, the intent generation module of the sending end generates a detail rendering intent for the video content. This intent explicitly includes detail rendering standards and limitations on undesirable rendering effects for the video content. As mentioned above, the detail rendering standard sets the standard that the sending end expects the receiving end to achieve when performing detail enhancement processing on the video content. The limitations on undesirable rendering effects set the maximum allowable components for various artifacts.

[0137] The negotiation initiation module sends a negotiation request to the receiving end through a dedicated negotiation channel. This request fully includes the detailed presentation intent and metadata generated above, ensuring that the receiving end fully understands the sending end's needs and auxiliary information. Upon receiving the negotiation request, the receiving end's negotiation response module responds based on its own capability description information, feeding back the capability description information to the sending end's negotiation initiation module.

[0138] After the sender and receiver reach an agreement, the negotiation initiation module generates a detailed presentation protocol and synchronizes it to the receiver via the negotiation channel. Once the receiver's negotiation response module confirms receipt, the detailed presentation protocol officially takes effect and serves as the core basis for subsequent detail enhancement processing, effect evaluation, and verification by the sender. Furthermore, to ensure the integrity and security of the protocol, contracts and capability configuration files can be hashed or signed. The evidence package can also include measurement context and traced signature logs to ensure trusted authentication.

[0139] In this embodiment, capability negotiation serves as the core coordination mechanism for achieving certifiable rich detail range (RDR). Its main components include three aspects: First, clarifying the criteria for RDR compliance, namely, the quantitative targets for detail rendering effects and the constraints on undesirable artifacts. Second, determining the achievable RDR level under current conditions by considering actual constraints such as device hardware performance and operating status. Third, standardizing appropriate measurement methods and evidence retention rules to ensure the evaluation process is repeatable and the results are auditable. Through this series of negotiation processes, a binding RDR contract is ultimately formed. This contract serves as the operating guideline for the closed-loop optimizer, effectively constraining the entire detail enhancement process and ensuring consistent and certifiable RDR experience across different devices and scenarios.

[0140] Compared to traditional video display pipelines, where differences in computing power, panel characteristics, motion processing algorithms, and firmware versions among devices, coupled with real-time constraints such as power limitations and thermal throttling, result in inconsistent quality of detail enhancement for the same content, making it difficult to establish a unified experience standard, the capability negotiation mechanism in this application transforms ambiguous enhancement logic into explicit protocol agreements, fundamentally solving the following problems: Establish capability awareness targets: Based on the actual capability configuration and real-time operating status of the receiving end, formulate reasonable RDR targets to avoid unstable experience caused by targets exceeding the device's capacity.

[0141] To achieve standardized compliance: The measurement conditions (such as sampling rules and aggregation windows) and quantification thresholds (such as the lower limit of spatial detail score and the upper limit of artifact penalty) for detailed presentation are clearly defined in the agreement in advance to ensure that there is a unified basis for compliance judgment.

[0142] Clearly define operational constraints: Clearly define the scope of operations that the optimizer can perform, as described below, including the allowed detailed enhancement processing modules, model versions, parameter tuning limits, etc., to avoid over-enhancement or ineffective operations.

[0143] The evidence content is specified: key data that needs to be recorded and reported during contract execution, including scoring results, parameter adjustment trajectories, equipment status snapshots, etc., to provide support for subsequent auditing and verification.

[0144] Through the above agreement, even when faced with heterogeneous devices and dynamic operating conditions, the same content intent can be transformed into a consistent and comparable RDR experience. At the same time, when device constraints change, a principle rollback can be performed according to the agreement to ensure the stability and continuity of the experience.

[0145] It should be noted that the measurement context is used to clarify the standardized structure of the conditions for calculating the RDR score, in order to eliminate ambiguity in the score and ensure that the same score is interpretable, comparable, and repeatable across different devices, content, or scenarios. For example, the measurement context may include at least one of the following: score version, sampling method (or sampling rule), ROI definition, scale level, time window, enabling artifact detectors, and normalization rule.

[0146] The scoring version is a unique identifier for the scoring algorithm and associated calibration parameters used to calculate the RDR score. During the negotiation phase, the scoring versions supported by the sender and receiver must be confirmed. The RDR contract will explicitly specify the applicable versions; failure to support a version will trigger a rollback or authentication rejection. The Return on Investment (ROI) definition clearly defines the classification criteria, partitioning methods, and weighting rules for the Region of Interest (ROI) to ensure consistent judgment of important regions across different devices and avoid score deviations due to differences in ROI partitioning. The time window refers to the frame sequence range for calculating temporal detail scores and temporal stability indicators. Temporal details (such as motion consistency and flicker prevention) must be based on continuous frame analysis; a clearly defined time window avoids score distortion caused by an analysis range that is too short or too long. Enabling the artifact detector clarifies the types of artifacts to be activated during RDR scoring and their corresponding detection algorithms, i.e., the unwanted artifacts used to calculate the artifact penalty (P_artifact). Normalization rules are used to unify the quantization range and calibration standards of RDR scores, ensuring that scores from different devices and with different content are on the same comparable dimension.

[0147] In some embodiments, depending on the needs of the actual application scenario, capability negotiation can adopt a variety of flexible participants and negotiation models: Direct negotiation mode: During the session initialization phase, the sending end (i.e., the video content provider) directly establishes communication with the receiving end. The two parties negotiate based on the device capability configuration file and RDR intent without the intervention of any intermediate links. It is suitable for simple point-to-point transmission scenarios.

[0148] Platform-based coordination and negotiation model: The platform acts as an intermediary, representing video content providers in centralized negotiations with multiple receiving devices. The platform is responsible for managing device-related policies, RDR hierarchy definitions, ecosystem governance rules, and standardized reporting formats. Through a unified negotiation framework, it efficiently adapts to the capability differences of different device types, reducing the negotiation complexity during large-scale deployments.

[0149] Multi-party negotiation model: Depending on the scenario requirements, additional parties can be introduced to jointly complete the negotiation. For example, SoC module providers can provide information on device capability constraints and supported model sets; TV OEM strategy layers can synchronize user-selected picture modes, power settings, and other preferences; operators can specify additional constraints such as bandwidth limits and reporting requirements. All parties collaborate to form a comprehensive RDR contract, ensuring the agreement meets multi-dimensional requirements. It should be understood that regardless of the negotiation model adopted, the final RDR contract must fully cover the main contents such as content intent, device capabilities, measurement standards, and evidentiary requirements to ensure the binding force and enforceability of the contract, laying the foundation for subsequent detail enhancements, scoring optimization, and certification verification.

[0150] The process of finalizing an RDR contract has been described above. After the RDR contract is finalized, it can be negotiated and adjusted according to the actual situation. The negotiation and adjustment process will be described below.

[0151] The receiving end also includes a status monitoring module. This module monitors the device's operating status and network transmission status in real time. Device operating status includes at least one of temperature, power consumption, and computing resource usage; network transmission status includes at least one of transmission rate, data integrity, and latency.

[0152] When the device operating constraints at the receiving end are detected to have changed based on the device operating status and network transmission status, and the protocol update trigger condition in the presentation details protocol is met, the negotiation response module is triggered to initiate a renegotiation request to the sending end to update the presentation details protocol.

[0153] In this embodiment, the receiver's status monitoring module senses the device's operation and network transmission dynamics in real time, ensuring that the detail rendering protocol matches the receiver's actual operating conditions. Regarding device operation status monitoring, the status monitoring module continuously collects data such as the receiver's temperature, power consumption, and computing resource usage. Temperature monitoring covers core components of the receiver, such as the chip (SoC / TCON) and display panel, tracking in real time whether thermal throttling trigger conditions occur. Power consumption monitoring accurately calculates the power consumption of the entire detail enhancement processing process, adapting to power consumption thresholds for different power modes such as normal, environmental, and turbo modes. Computational resource usage focuses on the utilization rate of computing units such as NPU / GPU / DSP, memory bandwidth usage, and the computational load allocation of each detail enhancement processing module.

[0154] In terms of network transmission status monitoring, the status monitoring module tracks transmission rate, data integrity, and latency. Transmission rate provides real-time feedback on the actual transmission bandwidth of the video stream and metadata, determining whether it meets the data stream supply requirements for detail enhancement. Data integrity is ensured by verifying indicators such as the hash value and retransmission rate of transmitted data packets, guaranteeing that video frames and metadata have not been lost or tampered with. Latency is recorded by tracking the end-to-end latency from data decoding at the receiving end to the completion of enhancement processing, verifying whether it meets the latency budget agreed upon in the detail rendering protocol.

[0155] The status monitoring module compares real-time collected device operating status and network transmission status data with the device operating constraints (such as upper temperature limit, power consumption threshold, lower limit of computing resource margin, minimum transmission rate requirement, upper latency limit, etc.) stipulated in the detailed presentation protocol. When a certain type of status data is detected to exceed the constraints stipulated in the protocol and meets the protocol update triggering conditions, the status monitoring module triggers the negotiation response module to initiate a renegotiation request to the sending end. For example, when the temperature at the receiving end continuously exceeds 60°C, triggering thermal throttling, causing the computing resource margin to drop below 30% (below the lower limit of 50% stipulated in the protocol), and this state continues for more than 3 video segments (the triggering duration stipulated in the protocol). Another example is when the network transmission rate drops below 1Mbps (below the 2Mbps required by the protocol), causing a metadata data transmission interruption exceeding 100ms (the latency threshold stipulated in the protocol), both of which will trigger renegotiation.

[0156] When the negotiation response module initiates a renegotiation request, it synchronously updates the receiver's capability description information, using the changed device operating status (e.g., thermal throttling status, current power mode, remaining computing power) and network transmission status (e.g., actual transmission rate, latency data) as dynamic constraint information to inform the sender of the upper limit of the current supported detail rendering level. After receiving the request, the sender's negotiation initiation module combines the updated capability description information to re-analyze the matching degree between the detail rendering intent and the receiver's current capabilities. It adjusts the compliance requirements of the detail rendering protocol (e.g., reducing the target score band, adjusting ROI weight), the effect evaluation method (e.g., switching to a no-reference scoring mode), and the adjustment rules (e.g., optimizing the fallback ladder), generates an updated detail rendering protocol, and synchronizes it to the receiver to ensure that the receiver can still achieve compliant and stable detail rendering effects under the new operating constraints.

[0157] In some embodiments, the status monitoring module of the receiving end is also used to periodically check the status of the processing module and the model version of the receiving end. When the processing module fails or the model version needs to be updated, the negotiation response module is triggered to re-evaluate the capabilities of the receiving end according to the changes in the status of the processing module or the model version, so as to update the capability description information, and send the updated capability description information to the negotiation initiation module of the sending end, so that the negotiation initiation module can renegotiate the detailed presentation protocol with the negotiation response module of the receiving end based on the updated capability description information.

[0158] In this embodiment, the receiver's status monitoring module can also periodically check the status of the detail enhancement processing module and the model version to ensure that the receiver's capability description information is consistent with the actual available capabilities. Regarding the status detection of the detail enhancement processing module, the status monitoring module can scan all supported detail enhancement processing modules at a preset period to detect whether the module has startup failures, operational anomalies, insufficient computing power output, or other fault conditions. For example, if the denoising module cannot load the core algorithm normally, or if the sharpening module frequently experiences stuttering when processing frames, these are both determined to be module faults.

[0159] In terms of model version detection, the status monitoring module can compare the current model version corresponding to each detail enhancement processing module with the latest version pushed by the sender or platform (e.g., the current SR model on the receiver is V3.0, and the latest version is V3.2). At the same time, it verifies the compatibility between the model version and the receiver hardware and firmware. When an available updated version is detected that meets the compatibility requirements, or when the current model version has incompatibility issues due to firmware upgrades, it is determined that the model version needs to be updated.

[0160] When the status monitoring module detects a failure in the detail enhancement processing module or a need to update the model version, it triggers the negotiation response module to initiate a capability reassessment process. The negotiation response module can reassess the actual capability boundaries of the receiver based on the type and scope of the failure processing module, or the capability changes brought about by the model version update (such as increased computing power requirements, expanded supported resolutions, enhanced artifact suppression capabilities, etc.). Based on the reassessment results, the negotiation response module updates the capability description information. Static capability information includes updating the current available status of the detail enhancement processing module, the updated model set ID, and the version number. Dynamic constraint information adjusts the computational budget, memory bandwidth requirements, etc., to match the new model version.

[0161] Subsequently, the negotiation response module sends the updated capability description information to the sender through the negotiation channel. Upon receiving this information, the sender's negotiation initiation module, in conjunction with the updated capability description information, re-analyzes the match between the original detail rendering intent and the receiver's current actual capabilities to renegotiate the detail rendering protocol. If a detail enhancement module malfunctions, rendering some enhancement functions unavailable, it may be necessary to lower the target score threshold or adjust the ROI weight allocation (e.g., abandoning advanced detail enhancement for the background region). If a model version update brings capability improvements, compliance requirements can be appropriately optimized (e.g., increasing the target value of the spatial detail score) or the allowed operation set can be expanded. Finally, both parties reach an updated detail rendering protocol, which takes effect simultaneously, ensuring that the protocol always adapts to the receiver's actual capabilities and avoiding substandard detail rendering effects or resource waste due to module malfunctions or version differences.

[0162] In some embodiments, the receiver can also drive closed-loop optimization through RDR scores. The optimizer (or controller, optimization tuning module) iteratively selects processing actions, such as module selection, model switching, parameter tuning, ROI allocation, and scheduling, using feedback from the RDR scoring engine to meet the RDR objectives defined in the RDR contract, while satisfying constraints including real-time latency, computational budget, power / thermal limits, memory bandwidth, and artifact limits. For example, Figure 6 As shown, the optimization and adjustment module is used to adjust the content processing method based on the detail enhancement results after the content processing module performs detail enhancement processing on the video content sent by the sending end according to the detail rendering protocol, until the processed video content meets the requirements of the detail rendering protocol. Based on this, through the closed-loop adjustment mechanism driven by RDR score, it is ensured that the processed video content meets the detail rendering standards and restrictions on poor rendering effects stipulated in the detail rendering protocol, avoiding compliance issues caused by fixed parameter settings, and improving the stability of detail rendering quality.

[0163] Optionally, the above-mentioned optimization and adjustment module is specifically used for: Analyze the evaluation results to identify the target details that did not meet the standards. Adjust the processing method for the video content according to the target details and the adjustment rules in the detail presentation protocol, including: selecting suitable processing algorithms, adjusting processing parameters to improve the intensity of detail presentation in the video content, or optimizing the allocation of processing resources at the receiving end.

[0164] The video content was resumed using the adjusted processing method until the detailed presentation evaluation results generated by the effect evaluation module met the compliance requirements of the detailed presentation protocol.

[0165] Record each adjustment to the handling method for a specific target item, the basis for the adjustment, and the changes in evaluation results to form an adjustment log, which serves as part of the certification credentials.

[0166] In this embodiment, the optimization and adjustment module, as a closed-loop control unit for achieving compliance with the detail rendering protocol at the receiving end, can initiate a precise adjustment process after the content processing module completes the detail enhancement processing of the video content according to the detail rendering protocol. First, the optimization and adjustment module can deeply analyze the detail rendering evaluation results generated by the effect evaluation module, extracting detail rendering effect indicators, poor rendering effect indicators, and distribution information for each display area. It then compares this quantitative data with the compliance requirements in the detail rendering protocol one by one to accurately locate the non-compliant target detail items. For example, if the spatial detail score is lower than the protocol-defined threshold T_spatial, edge sharpness or texture integrity is determined to be a non-compliant item. If the artifact penalty exceeds the protocol upper limit T_artifact, the corresponding ringing, halo, or flicker artifacts are listed as key adjustment targets. If the temporal detail score does not meet the constraint requirements, motion consistency features are locked as optimization targets.

[0167] For details that do not meet the target, the optimization and adjustment module formulates an adjustment plan based on the adjustment rules in the detail rendering protocol, combined with the receiver's capability description information (such as supported detail enhancement processing modules, model version, parameter range, etc.) and device operating constraints. Regarding the selection of processing algorithms, the optimization and adjustment module can enable or disable detail enhancement processing modules such as super-resolution (SR), denoising (NR), deblocking, sharpening, tone mapping, and motion estimation and compensation (MEMC). It can also adjust the processing order (such as deblocking-SR or SR-deblocking), or select suitable models from different model series to match specific models for different ROI regions such as faces and text. In terms of processing parameter adjustment, it can finely adjust continuous or discrete parameters such as sharpening intensity, denoising intensity, deblocking intensity, and temporal smoothing coefficient. Additionally, it can control the internal model inference mode (such as accuracy, number of iterations, and sampling steps). In terms of resource allocation optimization, the optimization and adjustment module can define ROI levels and allocate different model complexities, select suitable tile sizes, overlaps, guard band widths, and blending strategies, adaptively allocate computational resources to the regions with the highest marginal score gains, determine reasonable update frequencies, perform heavier processing on keyframes and lightweight updates on intermediate frames, and selectively allocate resources for temporal smoothing or motion consistency refinement. Furthermore, it can dynamically adjust the usage strategy of metadata features (such as edges, motion, depth, etc.) to reduce dependence when metadata is missing or noisy.

[0168] After the adjustment plan is determined, the content processing module will use the updated processing method to re-enhance the details of the video content, and the effect evaluation module will simultaneously regenerate the detailed presentation evaluation results. The optimization and adjustment module continuously repeats the closed-loop iterative process of parsing the evaluation results, locating the non-compliant items, formulating and implementing the adjustment plan, and reprocessing and evaluating.

[0169] As an example, the iteration process can flexibly employ strategies such as rule-based and heuristic control, search-based optimization, learning-based optimization, or multi-objective constraint optimization. For instance, when the ringing penalty exceeds the limit, the sharpening intensity is reduced and halo suppression is enabled, switching to an artifact-safe model series. Another example is when spatial details are insufficient but artifacts are compliant, switching to a higher-capacity SR model and increasing the intensity of detail reconstruction. Yet another example is when the flicker penalty increases, temporal smoothing is strengthened and motion consistency constraints are tightened. Simultaneously, each candidate action undergoes delay feasibility, thermal feasibility, memory feasibility, contractual feasibility, and artifact safety checks to ensure that adjustments do not violate hard constraints. The closed-loop process continues until the detailed presentation evaluation results generated by the effect evaluation module meet the protocol compliance requirements (total score ≥ T_overall, sub-scores meet the standard, and artifact penalty ≤ upper limit), convergence is detected, the iteration budget is exhausted, or a scene transition triggers a reset. To avoid oscillations, the optimization adjustment module can also enforce stability constraints such as hysteresis, minimum model hold time, and parameter smoothing.

[0170] Throughout the closed-loop adjustment process, the optimization and adjustment module can record in detail the adjustments made to each target detail (including specific information such as module selection, model switching, parameter changes, and resource allocation optimization), the basis for the adjustments (including unmet evaluation indicators and constraint status), and changes in evaluation results (including score vectors and compliance status comparisons before and after the adjustments), forming a complete adjustment log. This adjustment log integrates information such as initial configuration, contract parameters, constraint status snapshots, termination reasons and pipelines, and model version fingerprints, and synchronizes them to the credential generation module as an important component of the authentication credentials. This provides traceable and verifiable process evidence for subsequent verification at the sending end and platform auditing, ensuring the compliance and credibility of the detailed enhancement processing.

[0171] In some embodiments, the credential generation module can also perform local evidence recording, continuously recording key data during the enhancement process, including parameter configurations for each enhancement process, output results at each stage, the trajectory of changes in the detailed presentation evaluation results, specific actions and triggering reasons for optimization adjustments, and periodically capturing configuration snapshots of the processing flow to form a complete chain of evidence. This recorded data will serve as the core basis for the credential generation module to generate authentication credentials, ensuring that the sending end can trace the compliance of the entire processing process during verification. For example, the credential generation module is specifically used to: collect key data during the processing, including the detailed presentation protocol, content processing method, adjustment logs, detailed presentation evaluation results for each iteration, and device operating status data at the receiving end.

[0172] The collected key data is sorted and filtered, and the target data related to authentication is retained according to the detailed presentation protocol, while redundant data is removed.

[0173] The target data is linked and integrated with the final detailed presentation evaluation results to generate authentication credentials and credential identifiers.

[0174] Add a timestamp and receiver identifier to the authentication credentials, and send the authentication credentials and the final detailed evaluation results to the sender based on the receiver identifier.

[0175] The timestamp records the time the authentication credential was generated, and the receiver identifier is used to distinguish different receiving devices. Integrity processing can also be performed on the authentication credential to ensure that its content is not tampered with, thus forming the final authentication credential.

[0176] In this embodiment, the credential generation module, as the core data integration unit of the receiver authentication process, needs to comprehensively collect key data throughout the entire detail enhancement processing process to ensure the integrity and traceability of the authentication credentials. The collected key data mainly includes five types: First, the detail presentation protocol (i.e., the RDR contract), which includes the protocol ID, compliance objectives (such as target score bands, artifact caps, ROI weighting rules), measurement conditions (such as sampling rules, aggregation windows), allowed operation sets, and fallback steps, serving as the core basis for authentication. Second, the content processing method, recording the detail enhancement processing module used, key parameter configurations (such as sharpening intensity), and ROI resource allocation strategies. Third, the adjustment log, which includes the specific content of each optimization adjustment (such as module switching, parameter changes, resource allocation adjustments), the basis for the adjustment (corresponding to the unmet evaluation indicators), changes in evaluation results (comparison of RDR score vectors before and after adjustment), and the convergence reason (such as meeting compliance requirements or budget exhaustion). Fourth, the detailed evaluation results (i.e., RDR scores or score vectors) for each iteration include the global total score, sub-score vectors, detailed scores for each ROI region, confidence level, and preliminary compliance assessment. Fifth, the receiving end's device operating status data includes dynamic constraint data such as temperature, power consumption, computing resource utilization, memory bandwidth utilization, power mode, and thermal status (whether throttling is in effect) during processing, recorded synchronously with timestamps.

[0177] After data collection is complete, the credential generation module can organize and filter all key data according to the reporting requirements and evidence stipulated in the detailed presentation agreement. This ensures the precise retention of target data directly related to certification, such as RDR score vectors, parameter adjustment trajectories, device constraint snapshots, and core information like sampling and aggregation rules explicitly required to be recorded in the agreement. Simultaneously, redundant data is removed, including duplicate intermediate frame score records that fail to meet standards, temporary debugging logs unrelated to certification, and overly detailed device status parameters exceeding the agreement's requirements. This ensures the target data is concise and that no core information is omitted, meeting certification verification needs while controlling data volume.

[0178] Subsequently, the credential generation module deeply integrates the filtered target data with the final detailed presentation evaluation results. Using the detailed presentation protocol as the core index, it organizes content processing methods, adjustment logs, RDR scores for each instance, and device operating status data according to timeline and logical relationships, forming a structured data set. For example, it binds the sharpening parameter adjustment caused by excessive ringing artifacts to the corresponding RDR score change and the device thermal status data at that time, clearly presenting the complete chain of problem-adjustment-result-constraint. Based on this structured set, a formal authentication credential (including an RDR certificate and evidence package) is generated and a unique credential identifier (i.e., certificate ID) is assigned. In addition, this credential identifier is associated with the protocol ID, session ID, and segment ID for easy subsequent traceability and verification.

[0179] Finally, the credential generation module adds a precise timestamp and receiver identifier to the authentication credential. After completing the above processing, the module sends the authentication credential and detailed presentation evaluation results to the sender via in-band or side channel (i.e., the negotiation channel mentioned above) according to the communication routing rules corresponding to the receiver identifier. The authentication credential includes the target data structured set, credential identifier, timestamp, receiver identifier, and integrity hash value. The detailed presentation evaluation results clearly indicate the global RDR total score, score vector (S_spatial, S_temporal, P_artifact), confidence level, and compliance conclusion (pass, fail, or conditionally pass), providing a complete and standardized basis for the sender's verification work.

[0180] It should be noted that the aforementioned RDR certificate may include at least one of the following: identifier (such as certificate ID), compliance result, implemented RDR score, score vector, confidence level, measurement conditions, pipeline fingerprint (such as detail enhancement processing module and / or model version), and integrity proof (such as evidence package hash, digital signature, timestamp). The RDR evidence package is the detailed data supporting the RDR certificate claim, which may include at least one of the following: score output (such as the total RDR score, sub-scores, confidence level, etc. for each iteration), measurement context (such as scoring version, ROI weighting rules, sampling rules, etc.), optimizer trace summary (such as adjustment actions, iteration count, convergence reason), sampling proof (sampling frame, tile index, hash value), constraint state snapshot (such as latency, power, thermal state data), pipeline, and model version fingerprint. Of course, the data included in the evidence package, RDR certificate, RDR contract, etc., listed in this application are merely examples and can be configured according to requirements.

[0181] In addition, to avoid transmitting or storing the original video frames, the evidence package can retain privacy data such as hashes and sparse indexes. Sampled frames, ROI patching, and other evidence are hashed to obtain corresponding hash results. The receiving end stores or transmits the hash results instead of the original data, satisfying the integrity verification requirements during downstream validation while preventing the leakage of original content, such as the original video footage.

[0182] Sparse index tagging refers to using lightweight identifiers such as frame indexes and tile coordinates to replace complete data. For example, only the index combination of segment ID-sequence number-ROI category is recorded, instead of storing the entire frame image, which greatly reduces the risk of privacy leakage.

[0183] In some embodiments, after receiving the authentication credentials, the sending end can use the authentication credentials to verify the detail enhancement processing performed by the receiving end. The verification module of the sending end is specifically used for: The authentication credentials sent by the receiving end are parsed to extract the processing method, adjustment logs, and detailed presentation evaluation results. Based on the compliance requirements in the detailed presentation protocol and the authentication credentials, the detailed enhancement operation is verified to obtain a verification conclusion, including whether the verification passed or failed, and the verification conclusion is fed back to the receiving end.

[0184] In this embodiment, the verification module at the sending end verifies the compliance of the detail enhancement operation at the receiving end by parsing and verifying the authentication credentials sent by the receiving end. First, the verification module performs structured parsing of the authentication credentials to accurately extract key information. Then, after parsing, the verification module uses the compliance requirements in the detail presentation protocol as the core basis to perform verification of the detail enhancement operation from different dimensions. For example, verifying the compliance of the processing method: checking whether the extracted enhancement module and model version are within the allowed operation set stipulated in the protocol, whether the parameter configuration does not exceed the parameter limits stipulated in the protocol, and whether the ROI resource allocation strategy conforms to the ROI weighting rules in the protocol. Another example is verifying the compliance of the detail presentation effect: comparing the extracted RDR score with the compliance threshold stipulated in the protocol, and simultaneously judging the reliability of the evaluation result based on the confidence level. Yet another example is verifying the compliance of the adjustment process: checking the adjustment log to see if the receiving end executed the adjustment rules stipulated in the protocol for the non-compliant detail items, whether the adjustment trajectory is reasonable, and whether there are any operations that violate the fallback step.

[0185] Subsequently, based on the verification results across different dimensions, the verification module generates corresponding verification conclusions, categorized as either "verification passed" or "verification failed." If all verification dimensions meet the requirements of the detail rendering protocol, the verification is considered passed. If there are issues such as processing methods exceeding the allowed range, RDR scores falling below the threshold, artifact penalties exceeding limits, or violations in process adjustments, the verification is considered failed, and the violation type is noted in the conclusion (e.g., artifact penalties exceeding the protocol limit, use of an unauthorized detail enhancement processing module). After generating the verification conclusions, the verification module feeds them back to the receiving end through a negotiation channel, providing a basis for subsequent adjustments or renegotiations by the receiving end.

[0186] Optionally, the verification module at the sending end can also verify the integrity and authenticity of the authentication credentials, confirming that the content of the credentials has not been tampered with and that the source is reliable.

[0187] For integrity verification, the verification module can extract integrity proof fields from the authentication credential, including the evidence package hash value, digital signature, and timestamp. The verification module compares the hash value of the locally calculated core data of the authentication credential (such as processing method digest, RDR score vector, device class ID, etc.) with the hash value of the evidence package embedded in the authentication credential, and verifies the validity of the digital signature, confirming that the signature was issued by the agreed-upon key corresponding to the recipient's identifier, and that the signature content is consistent with the core data of the credential. If the hash value matches, the digital signature verification passes, and the timestamp is within a reasonable and valid range (not exceeding the session validity period or certificate expiration time), the authentication credential is deemed complete. If there are hash value mismatches, invalid signatures, or abnormal timestamps, the credential integrity is deemed compromised, and the verification conclusion is failure.

[0188] Regarding authenticity, the verification module can correlate details with the receiver identification rules stipulated in the protocol to verify whether the receiver identifier in the credential belongs to the scope of devices recognized in the protocol, thus ruling out the possibility of forged credentials by illegal devices. Furthermore, the verification module checks whether the pipeline fingerprint in the authentication credential is consistent with the capability description information determined by the receiver during the capability negotiation phase. If the model version or module combination recorded in the authentication credential exceeds the supported range declared by the receiver, the credential source is deemed unreliable. In addition, the verification module can also query historical interaction data recorded in the verification audit log to confirm whether the credential identifier and contract ID of the current authentication credential match previous negotiation records and data transmission trajectories, further verifying the legitimacy of the credential source.

[0189] Only after compliance verification is passed, integrity check is correct, and authenticity is confirmed to be reliable, will the verification module finally determine that the verification has passed. Otherwise, it will be determined that the verification has failed, and the abnormality type will be clearly marked in the verification conclusion (such as mismatched credential hash value, suspected tampering, receiver identifier not within the scope of protocol recognition, etc.), and it will be synchronously fed back to the receiver.

[0190] In some embodiments, such as Figure 7 As shown, the sending end also includes an audit log module, which is used to record the verification process and verification results of the verification module and generate a verification audit log.

[0191] In this embodiment, the audit log module at the sending end ensures the traceability and auditability of the verification process. It comprehensively records the entire verification process and final verification results, generating standardized verification audit logs. The logs mainly include three parts: First, basic verification information, including the timestamp of the verification request, the detailed presentation protocol ID, the credential identifier of the authentication certificate, the receiver identifier, the content, and the segment ID to clearly define the range of video content corresponding to the verification. Second, verification process details, including the parsing steps of the authentication certificate, the specific verification content of each verification dimension, any abnormal information discovered during the verification process (such as parameters approaching thresholds, low confidence levels, etc.), and the execution trajectory of the verification logic. Third, verification result information, including the final verification conclusion (pass or fail), the basis for the conclusion judgment (such as all compliance thresholds being met, artifact penalty exceeding the limit by 5 points), and the timestamp of the verification conclusion feedback.

[0192] The aforementioned verification audit logs can be generated in a structured format, ensuring clear and standardized log content for easy retrieval and verification. Verification audit logs can be linked to indexes of key data, such as the association identifiers of authentication credential parsing results and the specific reference locations of compliance requirements in detail presentation agreements. Furthermore, verification audit logs avoid recording raw video data or sensitive device information, ensuring privacy and security. Additionally, verification audit logs can be stored in append-only storage or tamper-proof log systems to prevent log content tampering and ensure their authenticity and audit validity. These verification audit logs can be used in subsequent dispute resolution, service level agreement (SLA) execution acceptance, compliance program audits, and other scenarios, providing auditors or certification entities with complete traceability of the verification process, while also providing data support for the sending end to optimize verification strategies and identify high-frequency violation types.

[0193] In some embodiments, the sending end may further include a dispute resolution module. The dispute resolution module is used to respond to dispute events during the verification process, retrieve the corresponding authentication credentials, detailed presentation evaluation results, and verification audit logs, and re-verify to obtain the re-verification result.

[0194] Send a request for supplementary evidence to the receiving end to obtain additional detailed data from the enhanced detail processing as supplementary evidence. Based on the verification results and supplementary evidence, make a final dispute resolution conclusion and synchronize it to the receiving end, and update the verification audit log.

[0195] In this embodiment, the dispute resolution module at the sending end is used to respond to disputes that arise during the verification process. Through a standardized review and verification process and supplementary evidence review, it makes a traceable final dispute resolution conclusion. Furthermore, scenarios triggering disputes include the receiving end raising objections to a failed verification conclusion, the discovery of incomplete or contradictory evidence during the verification process, and the auditing party questioning the verification results.

[0196] Upon triggering a dispute, the dispute resolution module first initiates a review and verification process: retrieving all data related to the dispute, such as the corresponding certification credentials, detailed assessment results (including RDR scores, score vectors, and / or relevant context), and generated verification audit logs. Based on this data and the RDR contract, the dispute resolution module re-executes the verification process according to independent review logic, checking for any omissions or misjudgments in the original verification process. This includes re-comparing the RDR score with compliance thresholds, verifying whether the review process complies with the agreement requirements, checking the completeness and reasonableness of the adjustment logs, and recalculating the verification score of the sampling anchor (e.g., evidence package containing sampling proof), ensuring the objectivity of the review results.

[0197] If insufficient data is found in the authentication credentials during the review process, the dispute resolution module can send a request for supplementary evidence to the receiving end. This request clearly specifies the required data type, such as more detailed parameter adjustment history records or a complete snapshot of the device's operating status. After the receiving end responds to the request and provides supplementary evidence, the dispute resolution module verifies the authenticity and relevance of the supplementary evidence (e.g., verifying data integrity through hash values ​​and confirming direct relevance to the disputed event) and integrates it into the review process.

[0198] Based on the verification results and supplementary evidence, the dispute resolution module makes a final dispute resolution conclusion. The conclusion types include maintaining the original verification conclusion, revising the original verification conclusion to "verification passed," or requiring the receiving end to re-perform the enhanced detail processing and submit new authentication credentials. The dispute resolution conclusion can detail the basis for the judgment, such as key findings during the verification process and the acceptance of supplementary evidence. After generating the dispute resolution conclusion, the dispute resolution module synchronizes the conclusion to the receiving end and updates the verification audit log to supplement the log with a complete record of the dispute resolution process (including the cause of the dispute, verification steps, details of supplementary evidence, final conclusion, and its basis), ensuring that the dispute resolution process is traceable and auditable.

[0199] In some embodiments, the authentication system also supports multi-level authentication granularity.

[0200] In some embodiments, the sending end includes a first session management module, and the receiving end includes a second session management module. Both the first session management module and the second session management module are used to manage the communication session between the sending end and the receiving end.

[0201] The first session management module and the second session management module negotiate to generate a unique session identifier, which is used to associate data transmission and processing operations during the session, and to include session-related information in authentication credentials and verification audit logs.

[0202] As an example, managing a communication session may include recording at least one of the following: session establishment time, duration, data transmission volume, and session status, including normal, abnormal, and terminated. When a session reaches a preset duration or an abnormal situation occurs, a session reset or re-establishment process is triggered to ensure communication stability and data integrity.

[0203] In this embodiment, the first session management module at the sending end and the second session management module at the receiving end are responsible for the full lifecycle management of the communication session, ensuring that the interaction process between the sending and receiving ends is traceable and that data transmission and processing operations are accurately correlated. During the session initialization phase (such as when a video playback session starts), the first and second session management modules establish a connection through a negotiation channel and negotiate to generate a unique session identifier based on preset identifier generation rules (such as combining device class ID, timestamp, and random sequence). Furthermore, the conference identifier adopts a standardized format to ensure no duplication in heterogeneous devices and multi-session concurrent scenarios.

[0204] After the session identifier is generated, it serves as the core link throughout the entire communication session. All data transmission and processing operations related to the session are bound to this identifier. For example, the detail rendering intent, metadata, and detail rendering protocol transmitted by the sender; the detail enhancement processing, RDR score (i.e., detail rendering evaluation result) calculation, optimization and adjustment actions, and authentication credential generation performed by the receiver; and the verification process and verification audit log recording by the sender—all these operations are associated with the session identifier, forming a traceable link between the session identifier and the entire process. For instance, the authentication credential generated by the receiver carries the corresponding session identifier, clearly identifying the specific communication session to which the authentication credential corresponds. The verification audit log of the sender also associates the session identifier, clearly tracing the video content processing session corresponding to the verification operation.

[0205] The first and second session management modules can collect and organize session-related information, including session start timestamps, session end timestamps, participant identifiers (sender service ID, receiver device class ID, etc.), the negotiated detailed presentation protocol version, and key events during the session (such as renegotiation triggering and dispute resolution initiation). This session-related information is added to the authentication credentials generated by the receiver and the verification audit log generated by the sender. The authentication credentials use session-related information as a basic attribute to ensure that downstream verification can clearly identify the session context to which the credentials belong. The verification audit log stores session-related information in association with the verification process and results, providing a complete session background for subsequent auditing and dispute resolution.

[0206] During session termination (such as when video playback ends or network interruption occurs), the first and second session management modules will synchronously record the session termination status and reason, and add the termination information to the authentication credentials and verification audit logs, completing the information loop for the entire session lifecycle. If an abnormal session interruption occurs, the session identifier and associated operation records will still be retained, supporting subsequent troubleshooting and continuation of incomplete operations.

[0207] In some embodiments, the sender further includes a policy update module, which is used to optimize and adjust the detail presentation intent based on the detail presentation evaluation results and verification conclusions fed back by the receiver, and synchronize the optimized detail presentation intent to the receiver for renegotiating the detail presentation protocol.

[0208] In this embodiment, the policy update module at the sending end is used to dynamically optimize the rendering intent and improve the consistency of experience across sessions. The measurement update module continuously optimizes and adjusts the rendering intent based on the rendering evaluation results fed back by the receiving end and the verification conclusions output by the verification module, and synchronizes it to the receiving end to support the renegotiation of the rendering protocol.

[0209] For example, if the verification result is successful and the RDR score is higher than the target threshold specified by the detail rendering protocol (e.g., higher than the target score with an upper limit of 10%), the detail rendering standard can be improved. If the verification result is unsuccessful and the reason for the failure is insufficient receiver capability (e.g., low-computing-power devices cannot meet the current spatial detail target), the optimization is adjusted to a capability-aware target, reducing the compliance threshold of the corresponding dimension (e.g., lowering the spatial detail score target from 85 points to 80 points), or adjusting the weight preference between detail and artifacts (switching from detail-first to a balanced mode).

[0210] Optionally, the strategy update module can also perform long-term optimization by combining multi-session feedback data. When multiple receivers report similar issues or verification results show that the standards are generally not met, the strategy update module analyzes the reasons, adjusts the detail presentation standards or restrictions on poor presentation effects, and generates optimized detail presentation intents. For batch feedback from the same type of device or the same content category, if it is found that the temporal detail scores of a specific type of video (such as sports events) are generally low, the strategy update module can strengthen motion-related detail requirements in the detail presentation intent (such as increasing the temporal detail score target and adding optimization prompts for motion consistency refinement). If it is detected that the receiver's metadata data transmission is incomplete due to network bandwidth limitations, metadata budget adaptation rules can be added to the detail presentation intent to clarify the fallback steps when bandwidth is insufficient.

[0211] After receiving the optimized presentation intent, the sending end can synchronize the optimized presentation intent to the receiving end through a negotiation channel. Upon receiving the response, the receiving end's negotiation module will combine its latest capability description information with the sending end's negotiation initiation module to restart the negotiation process for the presentation protocol. This will ultimately form a new protocol adapted to the current device capabilities, content characteristics, and network conditions, achieving a closed loop of feedback-optimization-negotiation and continuously improving the compliance rate of presentation effects and the stability of user experience.

[0212] In some embodiments, the integrity of data transmitted between the sender and receiver can be enforced by signature or hash to prevent tampering.

[0213] In some embodiments, the RDR contract retention policy defines the content that can be stored on the device or shared upstream.

[0214] In some embodiments, the authentication system can perform authentication at different granularities according to the actual application scenario requirements, to adapt to differences in storage budget, reporting requirements, and traceability levels, ensuring that the authentication mechanism achieves a balance between practicality and resource consumption. Authentication may include at least one of the following granularities: Each segment authentication: Authentication is performed based on video streaming media blocks, scenes, or time segments. For example, authentication credentials are generated for each segment. Each segment authentication focuses on the compliance of detailed presentation of a single segment, accurately reflecting the processing quality of different content segments, and is suitable for streaming media distribution scenarios with high real-time requirements.

[0215] Per-session authentication: This method integrates authentication data from all segments within a complete playback session for comprehensive evaluation, resulting in the authentication credential corresponding to that session. Per-session authentication is suitable for quality assurance scenarios involving complete video content, such as movie and TV series playback on long-form video platforms, and complete course playback in educational videos. Session-level authentication comprehensively assesses the overall compliance of a single viewing experience, providing content service providers with closed-loop data on the quality of a user's single viewing session, while simultaneously reducing frequent authentication data transmission and lowering system interaction overhead.

[0216] Authentication is performed by title / device category, using video title or device category as the aggregation unit.

[0217] Each mode is certified separately: certification is performed for different working modes of the receiver (such as ecosystem mode, standard mode, cinema mode, game mode, etc.) to ensure that the detailed presentation effect of each mode meets the design expectations, provide users with a quality basis for mode selection, and help device manufacturers optimize the parameter configuration of different modes.

[0218] In some embodiments, the sender can also accurately identify anomalies in detail presentation and system malfunctions by continuously aggregating and analyzing standardized report data (such as RDR scores, compliance rates, artifact penalties, negotiation records, etc.) reported by all receivers. Furthermore, the sender can trigger standardized remedial measures on the receiver to ensure the stability and compliance of the RDR experience across devices and scenarios. Anomaly types can include at least one of the following: score drift, unstable peaks, artifact regression, and constraint transitions. Score drift: Continuously declining compliance of device category or firmware version. Unstable peaks: Increased flicker penalties or frequent renegotiation. Artifact regression: Halo / ringing issues introduced by model updates. Constraint transitions: Thermal throttling due to environmental conditions or new power regulations.

[0219] Remedial measures may include at least one of the following: complete breach of contract or backup ladder, roll back model version, tighten the set of permitted actions, increase audit sampling of affected queues, or release targeted firmware patches.

[0220] It should be noted that the operations performed by the above module are merely examples, and this application does not impose any restrictions on the specific operations performed by the module. In general, the operations performed by the module are actually performed by the device to which the module belongs (such as the receiving end or the transmitting end).

[0221] In some embodiments, this application also provides a receiving end. The receiving end includes: The content processing module is used to perform detail enhancement processing on the video content sent by the sending end in accordance with the determined detail rendering protocol; The effect evaluation module is used to evaluate the detail presentation effect of the video content after detail enhancement processing according to the effect evaluation method in the detail presentation protocol, and generate a detail presentation evaluation result. The credential generation module is used to collect relevant data during the detail enhancement process, combine the detail rendering protocol and the detail rendering evaluation result to generate an authentication credential, and send the authentication credential and the detail rendering evaluation result to the sending end so that the sending end can verify the authentication credential and the detail rendering evaluation result based on the detail rendering protocol.

[0222] The operations performed by the receiving end can be found in the previous section on the receiving end, and will not be repeated here.

[0223] In some embodiments, this application also provides a transmitting end, the transmitting end comprising: The verification module is used to receive the authentication credentials and detail rendering evaluation results sent by the receiving end after performing detail enhancement processing on the video content, and to verify the authentication credentials and detail rendering evaluation results based on the detail rendering protocol; wherein, the detail enhancement processing is performed on the video content sent by the sending end according to the established detail rendering protocol, and the detail rendering evaluation results are generated after evaluating the detail rendering effect of the video content after detail enhancement processing according to the effect evaluation method in the detail rendering protocol.

[0224] The operations performed by the sending end can be found in the previous section on the sending end, and will not be repeated here.

[0225] In some embodiments, according to a fourth aspect of this application, a rich detail-oriented authentication method is provided, such as... Figure 8 As shown, the method includes: S101. The receiving end performs detail enhancement processing on the video content sent by the sending end according to the established detail presentation protocol.

[0226] S102. The receiving end evaluates the detail rendering effect of the video content after detail enhancement processing according to the effect evaluation method in the detail rendering protocol, and generates a detail rendering evaluation result.

[0227] S103. The receiving end collects relevant data during the detail enhancement process, combines the detail presentation protocol and detail presentation evaluation results, generates authentication credentials, and sends the authentication credentials and detail presentation evaluation results to the sending end.

[0228] S104. The sending end receives the authentication credentials and the detailed presentation evaluation results sent by the receiving end after performing detail enhancement processing on the video content.

[0229] S105. The sending end verifies the authentication credentials and the presentation details evaluation results based on the presentation details protocol.

[0230] The implementation process of S101-S105 can be referred to the relevant content about the receiver and sender in the authentication system above. In addition, for some other possible implementation sub-steps of this authentication method, please refer to the specific embodiments of the aforementioned authentication system, which will not be repeated here.

[0231] For example, in order to implement the functions of the modules of the aforementioned receiving end and transmitting end, or the aforementioned methods, both the receiving end and the transmitting end can include a processor and a memory. The aforementioned functional modules can be software functional modules or hardware modules. Taking a software functional module as an example, it can contain computer-executable instructions stored in the memory, which, when executed by the processor, can implement the aforementioned methods or the functions implemented by the corresponding functional modules.

[0232] It will be understood by those skilled in the art that any references to memory, storage, database, or other media used in the embodiments provided in this disclosure may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchlink, DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0233] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0234] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0235] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0236] The embodiments, implementation methods, and related technical features of this application can be combined and substituted for each other without conflict.

[0237] The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application without departing from the scope of the technical solution of this application shall still fall within the scope of the technical solution of this application.

Claims

1. A rich detail authentication system, characterized in that, This includes the sending and receiving ends of the communication connection, where: The receiving end includes: The content processing module is used to perform detail enhancement processing on the video content sent by the sending end in accordance with the determined detail rendering protocol; The effect evaluation module is used to evaluate the detail presentation effect of the video content after detail enhancement processing according to the effect evaluation method in the detail presentation protocol, and generate a detail presentation evaluation result. The credential generation module is used to collect relevant data during the detail enhancement process, combine the detail presentation protocol and the detail presentation evaluation result to generate an authentication credential, and send the authentication credential and the detail presentation evaluation result to the sending end for authentication. The sending end includes: The verification module is used to receive the authentication credentials and detail rendering evaluation results sent by the receiving end after performing detail enhancement on the video content, and to verify the authentication credentials and detail rendering evaluation results based on the detail rendering protocol.

2. The authentication system according to claim 1, characterized in that, The transmitting end also includes: The intent generation module is used to generate detailed presentation intents, which include detailed presentation standards and restrictions on poor presentation effects for the video content. The negotiation initiation module is used to negotiate capabilities with the receiving end based on the detailed presentation intent and determine the detailed presentation protocol. The receiving end also includes: The negotiation response module is used to respond to the negotiation request of the sending end based on the capability description information of the receiving end, and to jointly negotiate with the sending end to determine the detailed presentation protocol; the capability description information includes supported content processing methods and device operating constraints.

3. The authentication system according to claim 2, characterized in that, The sending end also includes a metadata generation module, used to analyze the video content, extract scene structure clues of the video content, and generate metadata for instructing the sending end to perform detail enhancement processing on the video content; the metadata includes at least one of edge clues, motion clues, depth clues, and important region identifiers of the video content; The negotiation initiation module of the sending end is also used to send the metadata and the detail presentation intent to the receiving end together, for negotiating and determining the detail presentation protocol with the receiving end.

4. The authentication system according to claim 3, characterized in that, The negotiation initiation module is also used to send a negotiation request to the receiving end through a negotiation channel, the negotiation request including the detailed presentation intent and the metadata; The negotiation response module is used to respond to the negotiation request and feed back the capability description information to the negotiation initiation module; The negotiation initiation module is also used to analyze the degree of matching between the detailed presentation intent and the capability description information, determine the level of detailed presentation that the receiving end can achieve under the current capability, exchange opinions with the receiving end, determine the compliance requirements, effect evaluation methods and adjustment rules of the detailed presentation protocol, generate the detailed presentation protocol, and synchronize the detailed presentation protocol to the receiving end.

5. The authentication system according to claim 2, characterized in that, The capability description information includes the static capability information and dynamic constraint information of the receiving end; The static capability information includes the resolution supported by the receiving end, refresh rate, processing module type for performing detail enhancement processing, and model version. The processing module type includes at least one of super-resolution processing module, denoising processing module, deblocking processing module, sharpening processing module, tone mapping processing module, and motion processing module. The dynamic constraint information includes at least one of the following: the current thermal state of the receiving end, power mode, computational margin, and memory bandwidth. The negotiation response module is further configured to integrate the static capability information and the dynamic constraint information to form the capability description information and send it to the sending end for capability negotiation with the sending end.

6. The authentication system according to claim 5, characterized in that, The receiving end also includes a status monitoring module, used for: The device operating status and network transmission status of the receiving end are monitored in real time. The device operating status includes at least one of temperature, power consumption, and computing resource usage. The network transmission status includes at least one of transmission rate, data integrity, and latency. When the device operating constraints of the receiving end are detected to have changed based on the device operating status and network transmission status, and the protocol update trigger condition in the detail presentation protocol is met, the negotiation response module is triggered to initiate a renegotiation request to the sending end to update the detail presentation protocol.

7. The authentication system according to claim 6, characterized in that, The receiving end's status monitoring module is also used to periodically check the status of the receiving end's processing module and model version. When the processing module malfunctions or the model version needs to be updated, the negotiation response module is triggered to re-evaluate the receiving end's capabilities based on the changes in the processing module's status or model version, so as to update the capability description information. The updated capability description information is then sent to the sending end's negotiation initiation module, enabling the negotiation initiation module to renegotiate the detailed presentation protocol with the receiving end's negotiation response module based on the updated capability description information.

8. The authentication system according to claim 1, characterized in that, The detail enhancement processing performed by the content processing module includes at least one of spatial detail enhancement operation, temporal detail stabilization operation, and poor presentation effect suppression operation.

9. The authentication system according to claim 1, characterized in that, The effect evaluation module is specifically used for: According to the evaluation method in the aforementioned detail rendering protocol, image data of the video content after detail enhancement processing is collected; Extract detail rendering-related features from the image data, wherein the detail rendering-related features include at least one of edge sharpness features, texture distribution features, color accuracy features, and motion consistency features; Based on the detail rendering standards in the detail rendering protocol, the detail rendering-related features are quantitatively evaluated to obtain detail rendering effect indicators; The image data is detected to produce undesirable presentation effects. The type and degree of the detected undesirable presentation effects are identified and quantified to obtain an undesirable presentation effect index. By combining the aforementioned detail rendering effect index and the aforementioned poor rendering effect index, the detail rendering evaluation result is generated. The detail rendering evaluation result includes the detail rendering status of each display area and the distribution of poor rendering effects.

10. The authentication system according to claim 9, characterized in that, The process by which the effect evaluation module extracts the relevant features in the detailed presentation includes: An edge detection algorithm is used to extract the edge information of the displayed object in the image data to generate edge sharpness features; The texture distribution features in the image data are extracted using a texture analysis algorithm. The texture distribution features include at least one of texture density, texture direction, and texture repetition pattern. By comparing the color information of the video content after detail enhancement with the color information of the original video content before detail enhancement, the degree of color deviation is calculated and color accuracy features are generated. Based on inter-frame motion analysis, the details of the displayed object are tracked in consecutive frames, the consistency of details during motion is evaluated, and motion consistency features are generated. The edge sharpness feature, texture distribution feature, color accuracy feature, and motion consistency feature are standardized and quantized to a unified dimension for generating the detail presentation evaluation result.

11. The authentication system according to claim 1, characterized in that, The receiving end also includes: The optimization and adjustment module is used to adjust the content processing method according to the detail rendering evaluation result after the content processing module performs detail enhancement processing on the video content sent by the sending end in accordance with the detail rendering protocol, until the processed video content meets the requirements of the detail rendering protocol.

12. The authentication system according to claim 11, characterized in that, The optimization and adjustment module is specifically used for: The analysis of these details presents the evaluation results, identifying the target details that did not meet the standards. Adjusting the processing method for the video content according to the target detail item and the adjustment rules in the detail presentation protocol includes: selecting an appropriate processing algorithm, adjusting processing parameters to improve the detail presentation intensity of the video content, or optimizing the processing resource allocation of the receiving end; The video content is then processed again using the adjusted method until the detail rendering evaluation result generated by the effect evaluation module meets the compliance requirements of the detail rendering protocol. Record each adjustment to the processing method for the aforementioned target detail, the basis for the adjustment, and the changes in evaluation results to form an adjustment log, which serves as a component of the certification credentials.

13. The authentication system according to claim 12, characterized in that, The voucher generation module is specifically used for: Collect key data during the processing, including the detailed presentation protocol, content processing method, adjustment log, evaluation results of each detailed presentation, and device operating status data of the receiving end; The collected key data is sorted and filtered, and the target data related to authentication is retained and redundant data is removed according to the detailed presentation protocol. The target data is linked and integrated with the final detailed presentation evaluation results to generate the authentication credential and the credential identifier of the authentication credential; Add a timestamp and a receiver identifier to the authentication credential, and send the authentication credential and the final detailed presentation evaluation result to the sender based on the receiver identifier.

14. The authentication system according to claim 13, characterized in that, The verification module is specifically used for: The authentication credentials sent by the receiving end are parsed, and the processing method, adjustment logs and details are extracted to present the evaluation results. Based on the compliance requirements in the presentation details protocol and the authentication credentials, the detail enhancement operation is verified to obtain a verification conclusion, which includes whether the verification is passed or failed, and the verification conclusion is fed back to the receiving end.

15. The authentication system according to claim 14, characterized in that, The sending end also includes an audit log module, which is used to record the verification process and verification results of the verification module and generate a verification audit log.

16. The authentication system according to claim 14, characterized in that, The sending end also includes a dispute handling module, which is used to respond to dispute events during the verification process, retrieve the corresponding authentication credentials, detailed presentation evaluation results and verification audit logs, and re-verify to obtain the verification result; Send a request for supplementary evidence to the receiving end to obtain other detailed data from the detail enhancement process as supplementary evidence; Based on the verification results and the supplementary evidence, a final dispute resolution conclusion is made and synchronized to the receiving end, and the verification audit log is updated.

17. The authentication system according to claim 16, characterized in that, The sending end includes a first session management module, and the receiving end includes a second session management module, used to manage the communication session between the sending end and the receiving end; The first session management module and the second session management module negotiate to generate a unique session identifier, which is used to associate data transmission and processing operations during the session, and to incorporate session-related information into the authentication credentials and the verification audit log.

18. The authentication system according to claim 1, characterized in that, The sending end also includes a policy update module, which is used to optimize and adjust the detail presentation intent based on the detail presentation evaluation results and verification conclusions fed back by the receiving end, and synchronize the optimized detail presentation intent to the receiving end for renegotiating the detail presentation protocol.

19. A receiving end, communicatively connected to a receiving end in the authentication system according to any one of claims 1 to 18, characterized in that, The receiving end includes: The content processing module is used to perform detail enhancement processing on the video content sent by the sending end in accordance with the determined detail rendering protocol; The effect evaluation module is used to evaluate the detail presentation effect of the video content after detail enhancement processing according to the effect evaluation method in the detail presentation protocol, and generate a detail presentation evaluation result. The credential generation module is used to collect relevant data during the detail enhancement process, combine the detail rendering protocol and the detail rendering evaluation result to generate an authentication credential, and send the authentication credential and the detail rendering evaluation result to the sending end so that the sending end can verify the authentication credential and the detail rendering evaluation result based on the detail rendering protocol.

20. The receiving end according to claim 19, characterized in that, The receiving end also includes: The negotiation response module is used to respond to the negotiation request of the sending end based on the capability description information of the receiving end, and to jointly negotiate with the sending end to determine the detailed presentation protocol; the capability description information includes supported content processing methods and device operating constraints.

21. The receiving end according to claim 20, characterized in that, The negotiation request is sent through a negotiation channel, and the negotiation request includes the detailed presentation intent and metadata; The negotiation response module is used to respond to the negotiation request, and feed back the capability description information to the sending end, so that the sending end can analyze the degree of matching between the detail presentation intent and the capability description information, determine the level of detail presentation that the receiving end can achieve under the current capability, exchange opinions with the receiving end, determine the compliance requirements, effect evaluation methods and adjustment rules of the detail presentation protocol, generate the detail presentation protocol, and synchronize the detail presentation protocol to the receiving end.

22. The receiving end according to claim 20, characterized in that, The capability description information includes the static capability information and dynamic constraint information of the receiving end; The static capability information includes the resolution supported by the receiving end, refresh rate, processing module type for performing detail enhancement processing, and model version. The processing module type includes at least one of super-resolution processing module, denoising processing module, deblocking processing module, sharpening processing module, tone mapping processing module, and motion processing module. The dynamic constraint information includes at least one of the following: the current thermal state of the receiving end, power mode, computational margin, and memory bandwidth. The negotiation response module is further configured to integrate the static capability information and the dynamic constraint information to form the capability description information and send it to the sending end for capability negotiation with the sending end.

23. The receiving end according to claim 22, characterized in that, The receiving end also includes a status monitoring module, used for: The device operating status and network transmission status of the receiving end are monitored in real time. The device operating status includes at least one of temperature, power consumption, and computing resource usage. The network transmission status includes at least one of transmission rate, data integrity, and latency. When the device operating constraints of the receiving end are detected to have changed based on the device operating status and network transmission status, and the protocol update trigger condition in the detail presentation protocol is met, the negotiation response module is triggered to initiate a renegotiation request to the sending end to update the detail presentation protocol.

24. The receiving end according to claim 23, characterized in that, The status monitoring module is also used to periodically check the status of the processing module and the model version of the receiving end. When the processing module malfunctions or the model version needs to be updated, the negotiation response module is triggered to re-evaluate the capabilities of the receiving end based on the changes in the status of the processing module or the model version, so as to update the capability description information and send the updated capability description information to the sending end, so that the sending end can renegotiate the detailed presentation protocol with the negotiation response module of the receiving end based on the updated capability description information.

25. The receiving end according to claim 19, characterized in that, The detail enhancement processing performed by the content processing module includes at least one of spatial detail enhancement operation, temporal detail stabilization operation, and poor presentation effect suppression operation.

26. The receiving end according to claim 19, characterized in that, The effect evaluation module is specifically used for: According to the evaluation method in the aforementioned detail rendering protocol, image data of the video content after detail enhancement processing is collected; Extract detail rendering-related features from the image data, wherein the detail rendering-related features include at least one of edge sharpness features, texture distribution features, color accuracy features, and motion consistency features; Based on the detail rendering standards in the detail rendering protocol, the detail rendering-related features are quantitatively evaluated to obtain detail rendering effect indicators; The image data is detected to produce undesirable presentation effects. The type and degree of the detected undesirable presentation effects are identified and quantified to obtain an undesirable presentation effect index. By combining the aforementioned detail rendering effect index and the aforementioned poor rendering effect index, the detail rendering evaluation result is generated. The detail rendering evaluation result includes the detail rendering status of each display area and the distribution of poor rendering effects.

27. The receiving end according to claim 26, characterized in that, The process by which the effect evaluation module extracts the relevant features in the detailed presentation includes: An edge detection algorithm is used to extract the edge information of the displayed object in the image data to generate edge sharpness features; The texture distribution features in the image data are extracted using a texture analysis algorithm. The texture distribution features include at least one of texture density, texture direction, and texture repetition pattern. By comparing the color information of the video content after detail enhancement with the color information of the original video content before detail enhancement, the degree of color deviation is calculated and color accuracy features are generated. Based on inter-frame motion analysis, the details of the displayed object are tracked in consecutive frames, the consistency of details during motion is evaluated, and motion consistency features are generated. The edge sharpness feature, texture distribution feature, color accuracy feature, and motion consistency feature are standardized and quantized to a unified dimension for generating the detail presentation evaluation result.

28. The receiving end according to claim 19, characterized in that, The receiving end also includes: The optimization and adjustment module is used to adjust the content processing method according to the detail rendering evaluation result after the content processing module performs detail enhancement processing on the video content sent by the sending end in accordance with the detail rendering protocol, until the processed video content meets the requirements of the detail rendering protocol.

29. The receiving end according to claim 28, characterized in that, The voucher generation module is specifically used for: Collect key data during the processing, including the detailed presentation protocol, content processing method, adjustment log, evaluation results of each detailed presentation, and device operating status data of the receiving end; The collected key data is sorted and filtered, and the target data related to authentication is retained and redundant data is removed according to the detailed presentation protocol. The target data is linked and integrated with the final detailed presentation evaluation results to generate the authentication credential and the credential identifier of the authentication credential; Add a timestamp and a receiver identifier to the authentication credential, and send the authentication credential and the final detailed presentation evaluation result to the sender based on the receiver identifier.

30. The receiving end according to claim 29, characterized in that, The receiving end also includes: The second session management module is used to negotiate with the sending end to generate a unique session identifier, which is used to associate data transmission and processing operations during the session; The second session management module is also used to incorporate session-related information into the authentication credentials.

31. A transmitting end, communicatively connected to a receiving end in the authentication system according to any one of claims 1 to 18, characterized in that, The sending end includes: The verification module is used to receive the authentication credentials and detail rendering evaluation results sent by the receiving end after performing detail enhancement processing on the video content, and to verify the authentication credentials and detail rendering evaluation results based on the detail rendering protocol; wherein, the detail enhancement processing is performed on the video content sent by the sending end according to the established detail rendering protocol, and the detail rendering evaluation results are generated after evaluating the detail rendering effect of the video content after detail enhancement processing according to the effect evaluation method in the detail rendering protocol.

32. The transmitting end according to claim 31, characterized in that, The transmitting end also includes: The intent generation module is used to generate detailed presentation intents, which include detailed presentation standards and restrictions on poor presentation effects for the video content. The negotiation initiation module is used to negotiate capabilities with the receiving end based on the detailed presentation intent and determine the detailed presentation protocol.

33. The transmitting end according to claim 32, characterized in that, The transmitting end also includes: The metadata generation module is used to analyze the video content, extract scene structure clues from the video content, and generate metadata to instruct the sending end to perform detail enhancement processing on the video content; the metadata includes at least one of edge clues, motion clues, depth clues, and important region identifiers of the video content; The negotiation initiation module is also used to send the metadata and the detail presentation intent to the receiving end together, and to negotiate with the receiving end to determine the detail presentation protocol.

34. The transmitting end according to claim 33, characterized in that, The negotiation initiation module is also used to send a negotiation request to the receiving end through a negotiation channel, the negotiation request including the detailed presentation intent and the metadata; The negotiation initiation module is also used to receive capability description information from the receiving end, analyze the degree of matching between the detailed presentation intent and the capability description information, determine the level of detailed presentation that the receiving end can achieve under the current capability, exchange opinions with the receiving end, determine the compliance requirements, effect evaluation methods and adjustment rules of the detailed presentation protocol, generate the detailed presentation protocol, and synchronize the detailed presentation protocol to the receiving end.

35. The transmitting end according to claim 31, characterized in that, The verification module is specifically used for: The authentication credentials sent by the receiving end are parsed, and the processing method, adjustment logs and details are extracted to present the evaluation results. Based on the compliance requirements in the presentation details protocol and the authentication credentials, the detail enhancement operation is verified to obtain a verification conclusion, which includes whether the verification is passed or failed, and the verification conclusion is fed back to the receiving end.

36. The transmitting end according to claim 35, characterized in that, The sending end also includes an audit log module, which is used to record the verification process and verification results of the verification module and generate a verification audit log.

37. The transmitting end according to claim 35, characterized in that, The sending end also includes a dispute handling module, which is used to respond to dispute events during the verification process, retrieve the corresponding authentication credentials, detailed presentation evaluation results and verification audit logs, and re-verify to obtain the verification result; Send a request for supplementary evidence to the receiving end to obtain other detailed data from the detail enhancement process as supplementary evidence; Based on the verification results and the supplementary evidence, a final dispute resolution conclusion is made and synchronized to the receiving end, and the verification audit log is updated.

38. The transmitting end according to claim 37, characterized in that, The transmitting end also includes: The first session management module is used to negotiate with the receiving end to generate a unique session identifier, which is used to associate data transmission and processing operations during the session. The first session management module is also used to incorporate session-related information into the authentication credentials and the verification audit log.

Images