IEEE Transactions on Broadcasting最新文献_第6页

Assessing AI-Generated Image Quality Using a Cross-Modal Hierarchical Perception Network 使用跨模态层次感知网络评估人工智能生成的图像质量

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2025-10-28 DOI: 10.1109/TBC.2025.3622417

Zhaoqing Pan;Yi Yang;Feng Yuan;Haoran Xie;Fu Lee Wang;Sam Kwong

{"title":"Assessing AI-Generated Image Quality Using a Cross-Modal Hierarchical Perception Network","authors":"Zhaoqing Pan;Yi Yang;Feng Yuan;Haoran Xie;Fu Lee Wang;Sam Kwong","doi":"10.1109/TBC.2025.3622417","DOIUrl":"https://doi.org/10.1109/TBC.2025.3622417","url":null,"abstract":"AI-Generated Images (AGIs) are increasingly used in various multimedia applications, making it essential to accurately assess the quality of AGIs to enhance user experience and optimize generative models. However, existing AI-Generated Image Quality Assessment (AGIQA) methods struggle to align fine-grained cross-modal semantics or capture diverse quality factors across multiple perceptual levels, limiting their effectiveness. To address these limitations, a Cross-modal Hierarchical Perception Network (CHPNet) is proposed for AGIQA, which simulates the hierarchical visual perception and adaptive decision-making mechanisms of the human brain. The proposed CHPNet comprises two key components: a Multi-level Cross-modal Interaction Network (MCINet) and an Adaptive Hierarchical Scoring Network (AHSNet). The MCINet is designed to generate multi-level quality-aware features by aligning and fusing visual and textual features at multiple semantic levels. To enhance semantic alignment, a Cross-modal Bidirectional Semantic Alignment Module (CBSAM) is built to improve the quality-aware feature extraction of MCINet by mitigating the semantic gap between cross-modal features. The AHSNet is developed to adaptively evaluate the importance of each perceptual level and assign importance-based weights to compute the final quality score. Extensive experiments on three AGIQA databases have demonstrated the effectiveness of the proposed CHPNet. The code of the proposed CHPNet is released at <uri>https://github.com/NUIST-Videocoding/CHPNet.git</uri>","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"291-302"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148513471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Face De-Identification: State-of-the-Art Methods and Comparative Studies 人脸去识别：最新方法与比较研究

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2025-12-11 DOI: 10.1109/TBC.2025.3639783

Jingyi Cao;Bo Liu;Xiangyi Chen;Ming Ding;Rong Xie;Li Song;Zhu Li;Wenjun Zhang;Yiyan Wu

{"title":"Face De-Identification: State-of-the-Art Methods and Comparative Studies","authors":"Jingyi Cao;Bo Liu;Xiangyi Chen;Ming Ding;Rong Xie;Li Song;Zhu Li;Wenjun Zhang;Yiyan Wu","doi":"10.1109/TBC.2025.3639783","DOIUrl":"https://doi.org/10.1109/TBC.2025.3639783","url":null,"abstract":"The widespread use of image acquisition technologies in broadcasting and multimedia services, together with advances in facial recognition, has raised serious privacy concerns. In live news or streaming interviews, audiences may legitimately view participants, yet automated systems can capture and recognize their identities without consent, creating significant legal and ethical risks. Face de-identification, which refers to the process of concealing or replacing personal identifiers, has therefore emerged as an effective means to protect the privacy of facial images. A significant number of methods for face de-identification have been proposed in recent years. In this survey, we provide a comprehensive review of state-of-the-art face de-identification methods, categorized into three levels: pixel-level, representation-level, and semantic-level techniques. We systematically evaluate these methods based on two key criteria, the effectiveness of privacy protection and preservation of image utility, highlighting their advantages and limitations. Our analysis includes qualitative and quantitative comparisons of the main algorithms, demonstrating that deep learning-based approaches, particularly those using Generative Adversarial Networks (GANs) and diffusion models, have achieved significant advancements in balancing privacy and utility. Experimental results reveal that while recent methods demonstrate strong privacy protection, trade-offs remain in visual fidelity and computational complexity. This survey not only summarizes the current landscape but also identifies key challenges and future research directions in face de-identification.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"260-280"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148513472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Learning Hierarchical and Spatial Semantic Correlations for Enhanced Multi-Scale Feature Compression 学习层次和空间语义关联增强多尺度特征压缩

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2025-12-11 DOI: 10.1109/TBC.2025.3639772

Liqian Zhang;Zhaoqing Pan;Li Li;Fu Lee Wang;Sam Kwong

{"title":"Learning Hierarchical and Spatial Semantic Correlations for Enhanced Multi-Scale Feature Compression","authors":"Liqian Zhang;Zhaoqing Pan;Li Li;Fu Lee Wang;Sam Kwong","doi":"10.1109/TBC.2025.3639772","DOIUrl":"https://doi.org/10.1109/TBC.2025.3639772","url":null,"abstract":"Multi-scale feature compression plays a crucial role in machine vision by reducing transmission and storage overhead while preserving task performance. However, existing methods neglect the distinct roles of features at different scales for recognizing objects of varying sizes and fail to fully capture the hierarchical semantic correlation across these scales. Moreover, they arrange feature representation into an image based on their original channel order and compress it using traditional codecs, which limits the exploitation of the spatial semantic correlation. To address these issues, a Hierarchical and Spatial Semantic Correlations (HSSC)-based multi-scale feature compression method is proposed in this paper, which jointly models hierarchical and spatial semantic correlations to improve compression efficiency. Specifically, to remove task-irrelevant redundancy in multi-scale features, a Semantic-guided Feature Transform Network (SFTNet) is developed, which leverages hierarchical semantic correlation to generate compact representations. In SFTNet, a semantic-guided cascaded fusion module is designed to reduce the data volume of multi-scale features by fusing task-relevant information based on intra- and inter-scale semantic correlations, while an adaptive feature reconstruction module is built to reconstruct multi-scale features by adaptively enhancing semantics for machine task analysis. To further eliminate redundancy in compact representations, a spatial semantic-based channel reordering strategy is proposed, which enhances spatial semantic correlation by reordering feature channels based on their local and global semantic similarities. Experimental results demonstrate that the proposed HSSC-based method outperforms state-of-the-art multi-scale feature compression methods.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"193-206"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148513270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Increasing Participant Quality of Experience in an Olfaction-Enhanced Live Educational Play 在嗅觉增强的现场教育游戏中提高参与者的体验质量

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2026-02-13 DOI: 10.1109/TBC.2026.3661399

Anderson Augusto Simiscuka;Mohammed Amine Togou;Gabriel-Miro Muntean

{"title":"Increasing Participant Quality of Experience in an Olfaction-Enhanced Live Educational Play","authors":"Anderson Augusto Simiscuka;Mohammed Amine Togou;Gabriel-Miro Muntean","doi":"10.1109/TBC.2026.3661399","DOIUrl":"https://doi.org/10.1109/TBC.2026.3661399","url":null,"abstract":"The recent advancements in Extended Reality (XR) have driven interest in multi-sensory experiences across creative and educational sectors. While prior research has explored inclusion of smell in immersive media and live performances, its application in live educational theater remains largely unexplored. This paper introduces MultiScent, a novel multi-sensory solution designed to enhance immersive experiences in theatrical performances via synchronized scent dispensers. By integrating olfactory stimuli with live performances, MultiScent aims to improve audience engagement and increase viewers’ perceived Quality of Experience (QoE). This study evaluates the MultiScent’s effectiveness by assessing the reactions of young audiences to an olfaction-enhanced educational play. To enhance viewers’ QoE, scents are used in sync with the play’s content using the innovative Olfactory Dispenser Synchronization Algorithm (ODSA). This study examines the effectiveness of using ODSA to maintain synchronization between sensory stimuli and play content and impact of scent-based multi-sensory integration on audience immersion and enjoyment. MultiScent utilizes a client-server architecture with multiple wireless-controlled scent dispensers. A MultiScent-based system was deployed in a live theatrical setting, where 70 young viewers attended a play that included synchronized olfactory elements. Post-performance interviews were conducted to assess the audience’s perception of immersion and QoE. The findings indicate that MultiScent significantly enhanced the audience’s enjoyment and engagement, demonstrating the potential of employing synchronized olfactory stimuli in theatrical environments. The results suggest that multi-sensory integration can contribute to a richer, more immersive experience, reinforcing the role of scent in enhancing emotional engagement in educational plays.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"281-290"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11395584","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148513799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Foreground-Aware Geometry Compression With Hybrid Attention for Large-Scale Point Clouds 基于混合关注的大尺度点云前景感知几何压缩

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2026-01-23 DOI: 10.1109/TBC.2026.3651190

Liang Xie;Haoran Li;Baoliang Chen;Ge Li;Sam Kwong;Wei Gao

{"title":"Foreground-Aware Geometry Compression With Hybrid Attention for Large-Scale Point Clouds","authors":"Liang Xie;Haoran Li;Baoliang Chen;Ge Li;Sam Kwong;Wei Gao","doi":"10.1109/TBC.2026.3651190","DOIUrl":"https://doi.org/10.1109/TBC.2026.3651190","url":null,"abstract":"Regions of Interest (ROI) play a crucial role in point cloud compression, especially in applications such as autonomous driving and robot navigation, where foreground regions often contain key information such as obstacles and object boundaries. However, traditional point cloud compression methods typically fail to optimize for these critical areas, instead applying uniform processing across the whole point clouds. The paper aims to optimize the point cloud compression process by allocating more bitstream resources to the foreground regions, thereby preserving important information in the point clouds. To achieve this, we propose a separation-based Foreground-Background Network (FB-Net) for compressing point cloud. The framework first identifies and separates the foreground and background regions, then designs an attention-based compression network, which includes multi-stage Occupancy Probability Estimation (OPE) module. The OPE module consist of an Attention-based Feature Extraction Layer (AFEL) and an Occupancy Probability Generation (AOPG) module. By controlling the number of OPE modules, we can allocate more bitstream resources to critical regions in the scene point cloud, such as tables and chairs, thus improving the performance of downstream detection tasks. Furthermore, to compensate for perceptual distortions in human vision, we design a large-scale receptive field-based Point Cloud Upsampling Network (PCU-Net), to enhance the objective quality. Through extensive experiments on point cloud datasets such as ScanNet and SUN RGB-D, we demonstrate that allocating more bitstream resources to the foreground regions benefits the accuracy of detection tasks. Compared to G-PCC and many state-of-the-art learning-based point cloud compression methods, our approach shows superior performance in detection tasks than compression-then-detection process methods.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"207-222"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IEEE Transactions on Broadcasting Publication Information IEEE广播出版信息汇刊

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2026-03-06 DOI: 10.1109/TBC.2026.3666582

引用次数: 0

Enabling Broadcasting-as-a-Service in Broadcast and Cellular Converged Networks 在广播和蜂窝融合网络中实现广播即服务

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2025-11-19 DOI: 10.1109/TBC.2025.3632562

Zihao Chen;Tao Lin;Yuting Gao;Yu Zhang;Yin Xu;Dazhi He;Jinyao Yan;Yahui Hu

{"title":"Enabling Broadcasting-as-a-Service in Broadcast and Cellular Converged Networks","authors":"Zihao Chen;Tao Lin;Yuting Gao;Yu Zhang;Yin Xu;Dazhi He;Jinyao Yan;Yahui Hu","doi":"10.1109/TBC.2025.3632562","DOIUrl":"https://doi.org/10.1109/TBC.2025.3632562","url":null,"abstract":"Broadcast and cellular converged networks (BCCNs) provide a promising solution to alleviate network congestion and improve user experience in environments such as massive-scale live streaming. However, existing studies on BCCNs mainly use a network-centric design paradigm where broadcast offloading is seen as just a network optimization tool, ignoring the interests of internet applications. As a result, this greatly limits their full potential. In this paper, we propose a novel Broadcast-as-a-Service framework for BCCNs, enabling broadcast network operators to provide dynamic on-demand broadcasting services to both internet applications and cellular network operators. Specifically, our proposed framework extends beyond traditional network-initiated offloading by allowing content providers to request broadcast offloading based on application-specific requirements. Furthermore, to enable dynamic on-demand broadcasting, we design a comprehensive metric to represent offloading benefits and formulate the offloading decision problem as an Integer Linear Programming (ILP), taking multi-stakeholder interests into account, including broadcast operators, cellular networks, and content providers. Last, we develop a Lyapunov-guided online offloading (LGO) algorithm to maximize offloading benefits while ensuring the long-term service violation rate does not exceed a predefined limit, avoiding frequent offloading switches. Extensive experiments validate that our proposed method outperforms all baselines in offloading utility while reducing the service violation rate by more than 10X compared to the best-performing baseline algorithm.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"2-13"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148510742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

LLM-Driven Multicast Network Slicing for 6G Non-Terrestrial Networks 基于llm驱动的6G非地面网络多播网络切片

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2025-12-18 DOI: 10.1109/TBC.2025.3637704

Zexuan Jing;Junsheng Mu;Yan Zhu;Lexi Xu;Fei Qi

{"title":"LLM-Driven Multicast Network Slicing for 6G Non-Terrestrial Networks","authors":"Zexuan Jing;Junsheng Mu;Yan Zhu;Lexi Xu;Fei Qi","doi":"10.1109/TBC.2025.3637704","DOIUrl":"https://doi.org/10.1109/TBC.2025.3637704","url":null,"abstract":"Non-terrestrial networks (NTNs) will play a central role in 6G broadcasting by providing ubiquitous coverage and the ability to offload media processing to space-borne compute. The next generation of immersive media services will require multicasting data from satellites to large groups of receivers with stringent latency and quality-of-service (QoS) demands. This paper proposes a large-language-model (LLM)-driven orchestration framework that translates broadcaster intents into optimized multicast network slices spanning ground and space resources. Building on the concept of computing-power networks (CPNs), the framework integrates compute-aware routing across GEO, MEO and LEO links, enforces delay-aware multicast constraints using an integer-linear-program formulation, and leverages LLM to adaptively admit multicast groups under resource constraints. A federated fine-tuning mechanism combined with a blockchain audit layer enables continuous improvement of the LLM while preserving data privacy and providing human-interpretable audit trails. Simulations demonstrate that the proposed method reduces end-to-end latency and energy consumption compared to traditional heuristics, while increasing the number of admitted multicast requests.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"29-45"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148511476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Playback Optimization for 360° Live Streaming via Bitrate Estimation and Download Scheduling 播放优化360°直播流通过比特率估计和下载调度

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2025-10-31 DOI: 10.1109/TBC.2025.3622364

Tao Lin;Zipeng Pan;Yuhong Xie;Yuan Zhang;Yin Xu

{"title":"Playback Optimization for 360° Live Streaming via Bitrate Estimation and Download Scheduling","authors":"Tao Lin;Zipeng Pan;Yuhong Xie;Yuan Zhang;Yin Xu","doi":"10.1109/TBC.2025.3622364","DOIUrl":"https://doi.org/10.1109/TBC.2025.3622364","url":null,"abstract":"The growing popularity of virtual reality (VR) applications and head-mounted displays (HMDs) has significantly increased the demand for 360-degree live streaming. However, delivering a premier immersive experience in such applications remains a challenge, primarily due to the stringent requirements of ultra-high bandwidth and ultra-low latency. Although several methods have been developed to address these problems, many overlook essential practical aspects of live streaming scenarios, such as significant bitrate fluctuations, vulnerability to rebuffering, and substantial computational demands. To overcome these limitations, we propose a novel playback optimization framework, LivePano, tailored for 360-degree live video streaming. Specifically, we first design an approach for bitrate estimation that utilizes our understanding of the spatio-temporal correlations found in the tiles of 360-degree videos, allowing us to accurately predict the actual bitrates of these tiles. Then, we develop a cascaded DRL-based technique to sequentially perform bitrate adaptation and download scheduling. This enables the active skipping of certain peripheral tiles to reduce playback stalls while minimizing the impact on the user experience. Finally, we employ imitation learning to distill these DRL models into computationally efficient decision trees, thereby accommodating the diverse computational capabilities of mobile devices. Extensive experimental results demonstrate that LivePano improves overall quality of experience (QoE) by 39.4% to 192.4% over the best-performing baseline algorithm across various network conditions, while maintaining minimal decision delay and CPU usage.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"156-169"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148513583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SFTNet: Spatial-Frequency Transformer Network for Learned Image Compression 用于学习图像压缩的空间频率变换器网络

IF 5.9 1区计算机科学

IEEE Transactions on Broadcasting Pub Date : 2026-03-01 Epub Date: 2025-12-03 DOI: 10.1109/TBC.2025.3635389

Meiqin Liu;Lingxue Li;Jiaming Liang;Chao Yao;Jian Jin;Tammam Tillo;Yao Zhao

{"title":"SFTNet: Spatial-Frequency Transformer Network for Learned Image Compression","authors":"Meiqin Liu;Lingxue Li;Jiaming Liang;Chao Yao;Jian Jin;Tammam Tillo;Yao Zhao","doi":"10.1109/TBC.2025.3635389","DOIUrl":"https://doi.org/10.1109/TBC.2025.3635389","url":null,"abstract":"In learned image compression (LIC), most existing methods process spatial domain information through convolutional neural networks (CNNs) or Transformers. However, they struggle to model frequency-domain correlations, particularly high-frequency detail correlations across regions. They suffer from high-frequency detail redundancy and inefficient uniform bit allocation, neither of which adapts to regional complexity. To address these issues, we propose the Spatial-Frequency Transformer Network (SFTNet) for LIC, which models cross-domain correlations between spatial structures and frequency components to enable region-adaptive bit allocation. To model these correlations, the Frequency-aware Transformer Block (FATB) employs dual attention mechanisms to collaboratively process spatial and frequency components. Specifically, its intra-block frequency attention enhances high-frequency details within each block by using low-frequency components as local structural guidance. Meanwhile, its inter-block spatial attention captures global consistency by modeling cross-block dependencies among low-frequency components. Building on this, the Feature Re-weighting Strategy (FRS) evaluates block importance via joint analysis of spatial dependencies and high-frequency energy. This enables dynamic alignment of the spatial-frequency features modeled by FATB for adaptive bit allocation, prioritizing structurally complex regions and compressing smooth areas to reduce redundancy. Experimental results demonstrate that our SFTNet outperforms VVC by −6.55% and −6.18% in BD-Rate on Kodak and CLIC datasets while achieving better reconstruction quality of the high-frequency details compared to recent LIC methods.","PeriodicalId":13159,"journal":{"name":"IEEE Transactions on Broadcasting","volume":"72 1","pages":"247-259"},"PeriodicalIF":5.9,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0