Weakly Supervised Semantic Segmentation(WSSS),which relies only on image-level labels,has attracted significant attention for its cost-effectiveness and scalability.Existing methods mainly enhance inter-class distinct...Weakly Supervised Semantic Segmentation(WSSS),which relies only on image-level labels,has attracted significant attention for its cost-effectiveness and scalability.Existing methods mainly enhance inter-class distinctions and employ data augmentation to mitigate semantic ambiguity and reduce spurious activations.However,they often neglect the complex contextual dependencies among image patches,resulting in incomplete local representations and limited segmentation accuracy.To address these issues,we propose the Context Patch Fusion with Class Token Enhancement(CPF-CTE)framework,which exploits contextual relations among patches to enrich feature repre-sentations and improve segmentation.At its core,the Contextual-Fusion Bidirectional Long Short-Term Memory(CF-BiLSTM)module captures spatial dependencies between patches and enables bidirectional information flow,yield-ing a more comprehensive understanding of spatial correlations.This strengthens feature learning and segmentation robustness.Moreover,we introduce learnable class tokens that dynamically encode and refine class-specific semantics,enhancing discriminative capability.By effectively integrating spatial and semantic cues,CPF-CTE produces richer and more accurate representations of image content.Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 validate that CPF-CTE consistently surpasses prior WSSS methods.展开更多
Succinonitrile(SN)-based polymer plastic crystal electrolytes(PPCEs)are regarded as promising candidates for lithium metal batteries but suffer from serious side reactions with Li metal.Herein,we propose a multi-dimen...Succinonitrile(SN)-based polymer plastic crystal electrolytes(PPCEs)are regarded as promising candidates for lithium metal batteries but suffer from serious side reactions with Li metal.Herein,we propose a multi-dimensional optimization strategy to alleviate the side reactions between SN and Li metal,and develop a highly stable poly-vinylethylene carbonate-based PPCE(PPCE-VEC).Moreover,we identify the intrinsic factors of multi-dimensional polymer structures on the electrolyte stability by three typical classes of polyesters.The PPCE-VEC constructed by in situ polymerization exhibits much better stability than poly-vinylene carbonate-based PPCE(PPCE-VCA)and poly-trifluoroethyl acrylate-based PPCE(PPCE-TFA),which is verified by its fewer SN-decomposition species in X-ray photoelectron spectroscopy(XPS)and outstanding full cell performance.The PPCE-VEC-enabled LiNi_(0.6)Co_(0.2)Mn_(0.2)O_(2)full cell achieve 73.7%capacity retention after 1400 cycles,which outperforms PPCE-VCA-and PPCE-TFA-enabled full cells(61.9%and 46.9%).Spectral analysis and theoretical calculation reveal that the high solvation ability of the carbonyl site,flexible polymer chain,and homogeneous electrolyte phase of PPCE-VEC are favorable to maximizing competition coordination with Li^(+)to weaken the Li^(+)–SN binding and shape an anion-rich solvation structure.This optimized polymer-involved Li^(+)solvation enhances SN stability and facilitates the formation of B/F enriched solid-electrolyte interphase(SEI),thus significantly improving PPCE stability.展开更多
In the age of big data,ensuring data privacy while enabling efficient encrypted data retrieval has become a critical challenge.Traditional searchable encryption schemes face difficulties in handling complex semantic q...In the age of big data,ensuring data privacy while enabling efficient encrypted data retrieval has become a critical challenge.Traditional searchable encryption schemes face difficulties in handling complex semantic queries.Additionally,they typically rely on honest but curious cloud servers,which introduces the risk of repudiation.Furthermore,the combined operations of search and verification increase system load,thereby reducing performance.Traditional verification mechanisms,which rely on complex hash constructions,suffer from low verification efficiency.To address these challenges,this paper proposes a blockchain-based contextual semantic-aware ciphertext retrieval scheme with efficient verification.Building on existing single and multi-keyword search methods,the scheme uses vector models to semantically train the dataset,enabling it to retain semantic information and achieve context-aware encrypted retrieval,significantly improving search accuracy.Additionally,a blockchain-based updatable master-slave chain storage model is designed,where the master chain stores encrypted keyword indexes and the slave chain stores verification information generated by zero-knowledge proofs,thus balancing system load while improving search and verification efficiency.Finally,an improved non-interactive zero-knowledge proof mechanism is introduced,reducing the computational complexity of verification and ensuring efficient validation of search results.Experimental results demonstrate that the proposed scheme offers stronger security,balanced overhead,and higher search verification efficiency.展开更多
This paper presents an intelligent patrol and security robot integrating 2D LiDAR and RGB-D vision sensors to achieve semantic simultaneous localization and mapping(SLAM),real-time object recognition,and dynamic obsta...This paper presents an intelligent patrol and security robot integrating 2D LiDAR and RGB-D vision sensors to achieve semantic simultaneous localization and mapping(SLAM),real-time object recognition,and dynamic obstacle avoidance.The system employs the YOLOv7 deep-learning framework for semantic detection and SLAM for localization and mapping,fusing geometric and visual data to build a high-fidelity 2D semantic map.This map enables the robot to identify and project object information for improved situational awareness.Experimental results show that object recognition reached 95.4%mAP@0.5.Semantic completeness increased from 68.7%(single view)to 94.1%(multi-view)with an average position error of 3.1 cm.During navigation,the robot achieved 98.0%reliability,avoided moving obstacles in 90.0%of encounters,and replanned paths in 0.42 s on average.The integration of LiDAR-based SLAMwith deep-learning–driven semantic perception establishes a robust foundation for intelligent,adaptive,and safe robotic navigation in dynamic environments.展开更多
Chinese abbreviations improve communicative efficiency by extracting key components from longer expressions.They are widely used in both daily communication and professional domains.However,existing abbreviation gener...Chinese abbreviations improve communicative efficiency by extracting key components from longer expressions.They are widely used in both daily communication and professional domains.However,existing abbreviation generation methods still face two major challenges.First,sequence-labeling-based approaches often neglect contextual meaning by making binary decisions at the character level,leading to abbreviations that fail to capture semantic completeness.Second,generation-basedmethods rely heavily on a single decoding process,which frequently produces correct abbreviations but ranks them lower due to inadequate semantic evaluation.To address these limitations,we propose a novel two-stage frameworkwithGeneration–Iterative Optimization forAbbreviation(GIOA).In the first stage,we design aChain-of-Thought prompting strategy and incorporate definitional and situational contexts to generate multiple abbreviation candidates.In the second stage,we introduce a Semantic Preservation Dynamic Adjustment mechanism that alternates between character-level importance estimation and semantic restoration to optimize candidate ranking.Experiments on two public benchmark datasets show that our method outperforms existing state-of-the-art approaches,achieving Hit@1 improvements of 15.15%and 13.01%,respectively,while maintaining consistent results in Hit@3.展开更多
High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes an...High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.展开更多
Advanced traffic monitoring systems encounter substantial challenges in vehicle detection and classification due to the limitations of conventional methods,which often demand extensive computational resources and stru...Advanced traffic monitoring systems encounter substantial challenges in vehicle detection and classification due to the limitations of conventional methods,which often demand extensive computational resources and struggle with diverse data acquisition techniques.This research presents a novel approach for vehicle classification and recognition in aerial image sequences,integrating multiple advanced techniques to enhance detection accuracy.The proposed model begins with preprocessing using Multiscale Retinex(MSR)to enhance image quality,followed by Expectation-Maximization(EM)Segmentation for precise foreground object identification.Vehicle detection is performed using the state-of-the-art YOLOv10 framework,while feature extraction incorporates Maximally Stable Extremal Regions(MSER),Dense Scale-Invariant Feature Transform(Dense SIFT),and Zernike Moments Features to capture distinct object characteristics.Feature optimization is further refined through a Hybrid Swarm-based Optimization algorithm,ensuring optimal feature selection for improved classification performance.The final classification is conducted using a Vision Transformer,leveraging its robust learning capabilities for enhanced accuracy.Experimental evaluations on benchmark datasets,including UAVDT and the Unmanned Aerial Vehicle Intruder Dataset(UAVID),demonstrate the superiority of the proposed approach,achieving an accuracy of 94.40%on UAVDT and 93.57%on UAVID.The results highlight the efficacy of the model in significantly enhancing vehicle detection and classification in aerial imagery,outperforming existing methodologies and offering a statistically validated improvement for intelligent traffic monitoring systems compared to existing approaches.展开更多
Weakly supervised semantic segmentation(WSSS)is a tricky task,which only provides category information for segmentation prediction.Thus,the key stage of WSSS is to generate the pseudo labels.For convolutional neural n...Weakly supervised semantic segmentation(WSSS)is a tricky task,which only provides category information for segmentation prediction.Thus,the key stage of WSSS is to generate the pseudo labels.For convolutional neural network(CNN)based methods,in which class activation mapping(CAM)is proposed to obtain the pseudo labels,and only concentrates on the most discriminative parts.Recently,transformer-based methods utilize attention map from the multi-headed self-attention(MHSA)module to predict pseudo labels,which usually contain obvious background noise and incoherent object area.To solve the above problems,we use the Conformer as our backbone,which is a parallel network based on convolutional neural network(CNN)and Transformer.The two branches generate pseudo labels and refine them independently,and can effectively combine the advantages of CNN and Transformer.However,the parallel structure is not close enough in the information communication.Thus,parallel structure can result in poor details about pseudo labels,and the background noise still exists.To alleviate this problem,we propose enhancing convolution CAM(ECCAM)model,which have three improved modules based on enhancing convolution,including deeper stem(DStem),convolutional feed-forward network(CFFN)and feature coupling unit with convolution(FCUConv).The ECCAM could make Conformer have tighter interaction between CNN and Transformer branches.After experimental verification,the improved modules we propose can help the network perceive more local information from images,making the final segmentation results more refined.Compared with similar architecture,our modules greatly improve the semantic segmentation performance and achieve70.2%mean intersection over union(mIoU)on the PASCAL VOC 2012 dataset.展开更多
In image analysis,high-precision semantic segmentation predominantly relies on supervised learning.Despite significant advancements driven by deep learning techniques,challenges such as class imbalance and dynamic per...In image analysis,high-precision semantic segmentation predominantly relies on supervised learning.Despite significant advancements driven by deep learning techniques,challenges such as class imbalance and dynamic performance evaluation persist.Traditional weighting methods,often based on pre-statistical class counting,tend to overemphasize certain classes while neglecting others,particularly rare sample categories.Approaches like focal loss and other rare-sample segmentation techniques introduce multiple hyperparameters that require manual tuning,leading to increased experimental costs due to their instability.This paper proposes a novel CAWASeg framework to address these limitations.Our approach leverages Grad-CAM technology to generate class activation maps,identifying key feature regions that the model focuses on during decision-making.We introduce a Comprehensive Segmentation Performance Score(CSPS)to dynamically evaluate model performance by converting these activation maps into pseudo mask and comparing them with Ground Truth.Additionally,we design two adaptive weights for each class:a Basic Weight(BW)and a Ratio Weight(RW),which the model adjusts during training based on real-time feedback.Extensive experiments on the COCO-Stuff,CityScapes,and ADE20k datasets demonstrate that our CAWASeg framework significantly improves segmentation performance for rare sample categories while enhancing overall segmentation accuracy.The proposed method offers a robust and efficient solution for addressing class imbalance in semantic segmentation tasks.展开更多
As a key node of modern transportation network,the informationization management of road tunnels is crucial to ensure the operation safety and traffic efficiency.However,the existing tunnel vehicle modeling methods ge...As a key node of modern transportation network,the informationization management of road tunnels is crucial to ensure the operation safety and traffic efficiency.However,the existing tunnel vehicle modeling methods generally have problems such as insufficient 3D scene description capability and low dynamic update efficiency,which are difficult to meet the demand of real-time accurate management.For this reason,this paper proposes a vehicle twin modeling method for road tunnels.This approach starts from the actual management needs,and supports multi-level dynamic modeling from vehicle type,size to color by constructing a vehicle model library that can be flexibly invoked;at the same time,semantic constraint rules with geometric layout,behavioral attributes,and spatial relationships are designed to ensure that the virtual model matches with the real model with a high degree of similarity;ultimately,the prototype system is constructed and the case region is selected for the case study,and the dynamic vehicle status in the tunnel is realized by integrating real-time monitoring data with semantic constraints for precise virtual-real mapping.Finally,the prototype system is constructed and case experiments are conducted in selected case areas,which are combined with real-time monitoring data to realize dynamic updating and three-dimensional visualization of vehicle states in tunnels.The experiments show that the proposed method can run smoothly with an average rendering efficiency of 17.70 ms while guaranteeing the modeling accuracy(composite similarity of 0.867),which significantly improves the real-time and intuitive tunnel management.The research results provide reliable technical support for intelligent operation and emergency response of road tunnels,and offer new ideas for digital twin modeling of complex scenes.展开更多
A composite anti-disturbance predictive control strategy employing a Multi-dimensional Taylor Network(MTN)is presented for unmanned systems subject to time-delay and multi-source disturbances.First,the multi-source di...A composite anti-disturbance predictive control strategy employing a Multi-dimensional Taylor Network(MTN)is presented for unmanned systems subject to time-delay and multi-source disturbances.First,the multi-source disturbances are addressed according to their specific characteristics as follows:(A)an MTN data-driven model,which is used for uncertainty description,is designed accompanied with the mechanism model to represent the unmanned systems;(B)an adaptive MTN filter is used to remove the influence of the internal disturbance;(C)an MTN disturbance observer is constructed to estimate and compensate for the influence of the external disturbance;(D)the Extended Kalman Filter(EKF)algorithm is utilized as the learning mechanism for MTNs.Second,to address the time-delay effect,a recursiveτstep-ahead MTN predictive model is designed utilizing recursive technology,aiming to mitigate the impact of time-delay,and the EKF algorithm is employed as its learning mechanism.Then,the MTN predictive control law is designed based on the quadratic performance index.By implementing the proposed composite controller to unmanned systems,simultaneous feedforward compensation and feedback suppression to the multi-source disturbances are conducted.Finally,the convergence of the MTN and the stability of the closed-loop system are established utilizing the Lyapunov theorem.Two exemplary applications of unmanned systems involving unmanned vehicle and rigid spacecraft are presented to validate the effectiveness of the proposed approach.展开更多
The Internet of Vehicles (IoV) has become an important direction in the field of intelligent transportation, in which vehicle positioning is a crucial part. SLAM (Simultaneous Localization and Mapping) technology play...The Internet of Vehicles (IoV) has become an important direction in the field of intelligent transportation, in which vehicle positioning is a crucial part. SLAM (Simultaneous Localization and Mapping) technology plays a crucial role in vehicle localization and navigation. Traditional Simultaneous Localization and Mapping (SLAM) systems are designed for use in static environments, and they can result in poor performance in terms of accuracy and robustness when used in dynamic environments where objects are in constant movement. To address this issue, a new real-time visual SLAM system called MG-SLAM has been developed. Based on ORB-SLAM2, MG-SLAM incorporates a dynamic target detection process that enables the detection of both known and unknown moving objects. In this process, a separate semantic segmentation thread is required to segment dynamic target instances, and the Mask R-CNN algorithm is applied on the Graphics Processing Unit (GPU) to accelerate segmentation. To reduce computational cost, only key frames are segmented to identify known dynamic objects. Additionally, a multi-view geometry method is adopted to detect unknown moving objects. The results demonstrate that MG-SLAM achieves higher precision, with an improvement from 0.2730 m to 0.0135 m in precision. Moreover, the processing time required by MG-SLAM is significantly reduced compared to other dynamic scene SLAM algorithms, which illustrates its efficacy in locating objects in dynamic scenes.展开更多
Ecological monitoring vehicles are equipped with a range of sensors and monitoring devices designed to gather data on ecological and environmental factors.These vehicles are crucial in various fields,including environ...Ecological monitoring vehicles are equipped with a range of sensors and monitoring devices designed to gather data on ecological and environmental factors.These vehicles are crucial in various fields,including environmental science research,ecological and environmental monitoring projects,disaster response,and emergency management.A key method employed in these vehicles for achieving high-precision positioning is LiDAR(lightlaser detection and ranging)-Visual Simultaneous Localization and Mapping(SLAM).However,maintaining highprecision localization in complex scenarios,such as degraded environments or when dynamic objects are present,remains a significant challenge.To address this issue,we integrate both semantic and texture information from LiDAR and cameras to enhance the robustness and efficiency of data registration.Specifically,semantic information simplifies the modeling of scene elements,reducing the reliance on dense point clouds,which can be less efficient.Meanwhile,visual texture information complements LiDAR-Visual localization by providing additional contextual details.By incorporating semantic and texture details frompaired images and point clouds,we significantly improve the quality of data association,thereby increasing the success rate of localization.This approach not only enhances the operational capabilities of ecological monitoring vehicles in complex environments but also contributes to improving the overall efficiency and effectiveness of ecological monitoring and environmental protection efforts.展开更多
Lower back pain is one of the most common medical problems in the world and it is experienced by a huge percentage of people everywhere.Due to its ability to produce a detailed view of the soft tissues,including the s...Lower back pain is one of the most common medical problems in the world and it is experienced by a huge percentage of people everywhere.Due to its ability to produce a detailed view of the soft tissues,including the spinal cord,nerves,intervertebral discs,and vertebrae,Magnetic Resonance Imaging is thought to be the most effective method for imaging the spine.The semantic segmentation of vertebrae plays a major role in the diagnostic process of lumbar diseases.It is difficult to semantically partition the vertebrae in Magnetic Resonance Images from the surrounding variety of tissues,including muscles,ligaments,and intervertebral discs.U-Net is a powerful deep-learning architecture to handle the challenges of medical image analysis tasks and achieves high segmentation accuracy.This work proposes a modified U-Net architecture namely MU-Net,consisting of the Meijering convolutional layer that incorporates the Meijering filter to perform the semantic segmentation of lumbar vertebrae L1 to L5 and sacral vertebra S1.Pseudo-colour mask images were generated and used as ground truth for training the model.The work has been carried out on 1312 images expanded from T1-weighted mid-sagittal MRI images of 515 patients in the Lumbar Spine MRI Dataset publicly available from Mendeley Data.The proposed MU-Net model for the semantic segmentation of the lumbar vertebrae gives better performance with 98.79%of pixel accuracy(PA),98.66%of dice similarity coefficient(DSC),97.36%of Jaccard coefficient,and 92.55%mean Intersection over Union(mean IoU)metrics using the mentioned dataset.展开更多
Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is cr...Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is crucial for computationally limited portable devices such as augmented reality and virtual reality.With the rapid advancements in deep learning,many network models have been developed specifically for eye image segmentation.Some methods divide the segmentation process into multiple stages to achieve model parameter miniaturization while enhancing output through post processing techniques to improve segmentation accuracy.These approaches significantly increase the inference time.Other networks adopt more complex encoding and decoding modules to achieve end-to-end output,which requires substantial computation.Therefore,balancing the model’s size,accuracy,and computational complexity is essential.To address these challenges,we propose a lightweight asymmetric UNet architecture and a projection loss function.We utilize ResNet-3 layer blocks to enhance feature extraction efficiency in the encoding stage.In the decoding stage,we employ regular convolutions and skip connections to upscale the feature maps from the latent space to the original image size,balancing the model size and segmentation accuracy.In addition,we leverage the geometric features of the eye region and design a projection loss function to further improve the segmentation accuracy without adding any additional inference computational cost.We validate our approach on the OpenEDS2019 dataset for virtual reality and achieve state-of-the-art performance with 95.33%mean intersection over union(mIoU).Our model has only 0.63M parameters and 350 FPS,which are 68%and 200%of the state-of-the-art model RITNet,respectively.展开更多
Semantic segmentation is a core task in computer vision that allows AI models to interact and understand their surrounding environment. Similarly to how humans subconsciously segment scenes, this ability is crucial fo...Semantic segmentation is a core task in computer vision that allows AI models to interact and understand their surrounding environment. Similarly to how humans subconsciously segment scenes, this ability is crucial for scene understanding. However, a challenge many semantic learning models face is the lack of data. Existing video datasets are limited to short, low-resolution videos that are not representative of real-world examples. Thus, one of our key contributions is a customized semantic segmentation version of the Walking Tours Dataset that features hour-long, high-resolution, real-world data from tours of different cities. Additionally, we evaluate the performance of open-vocabulary, semantic model OpenSeeD on our own custom dataset and discuss future implications.展开更多
Few-shot image semantic segmentation aims to achieve pixel-level classification for novel classes using only a few labeled examples.The method first trains the segmentation model on base classes,and then adapts it to ...Few-shot image semantic segmentation aims to achieve pixel-level classification for novel classes using only a few labeled examples.The method first trains the segmentation model on base classes,and then adapts it to novel classes.Although existing methods have achieved remarkable performance in few-shot image semantic segmentation,they still face the following challenges.Traditional methods typically rely on mask average pooling to generate single-category prototype vectors and perform feature matching via metric learning,but they exhibit significant limitations in modeling inter-category relationships and addressing complex background interference.Inspired by the analogy-based transfer mechanisms in cognitive psychology,we propose a Generalized Prototype Network(GPNet)to enhance the model's generalization ability for unseen categories and improve robustness in feature matching.GPNet consists of two key modules.The first is a generalized prototype enhancement module,which explores potential inter-category relationships to construct more discriminative category prototype representations.The second is a multi-scale feature alignment module,which dynamically aligns support and query features across multiple scales using an attention mechanism,thus mitigating background interference in complex scenarios.Experimental results demonstrate that the proposed method significantly outperforms existing state-of-the-art approaches on several few-shot semantic segmentation benchmarks,validating its effectiveness and generalization capabilities.展开更多
In recent years,deep learning-based semantic communications have shown great potential to enhance the performance of communication systems.This has led to the belief that semantic communications represent a breakthrou...In recent years,deep learning-based semantic communications have shown great potential to enhance the performance of communication systems.This has led to the belief that semantic communications represent a breakthrough beyond the Shannon paradigm and will play an essential role in future communications.To narrow the gap between current research and future vision,after an overview of semantic communications,this article presents and discusses ten fundamental and critical challenges in today’s semantic communication field.These challenges are divided into theory foundation,system design,and practical implementation.Challenges related to the theory foundation including semantic capacity,entropy,and rate-distortion are discussed first.Then,the system design challenges encompassing architecture,knowledge base,joint semantic-channel coding,tailored transmission scheme,and impairment are posed.The last two challenges associated with the practical implementation lie in cross-layer optimization for networks and standardization.For each challenge,efforts to date and thoughtful insights are provided.展开更多
With the rapid development of artificial intelligence and the Internet of Things,along with the growing demand for privacy-preserving transmission,the need for efficient and secure communication systems has become inc...With the rapid development of artificial intelligence and the Internet of Things,along with the growing demand for privacy-preserving transmission,the need for efficient and secure communication systems has become increasingly urgent.Traditional communication methods transmit data at the bit level without considering its semantic significance,leading to redundant transmission overhead and reduced efficiency.Semantic communication addresses this issue by extracting and transmitting only the mostmeaningful semantic information,thereby improving bandwidth efficiency.However,despite reducing the volume of data,it remains vulnerable to privacy risks,as semantic features may still expose sensitive information.To address this,we propose an entropy-bottleneck-based privacy protection mechanism for semantic communication.Our approach uses semantic segmentation to partition images into regions of interest(ROI)and regions of non-interest(RONI)based on the receiver’s needs,enabling differentiated semantic transmission.By focusing transmission on ROIs,bandwidth usage is optimized,and non-essential data is minimized.The entropy bottleneck model probabilistically encodes the semantic information into a compact bit stream,reducing correlation between the transmitted content and the original data,thus enhancing privacy protection.The proposed framework is systematically evaluated in terms of compression efficiency,semantic fidelity,and privacy preservation.Through comparative experiments with traditional and state-of-the-art methods,we demonstrate that the approach significantly reduces data transmission,maintains the quality of semantically important regions,and ensures robust privacy protection.展开更多
Remote driving,an emergent technology enabling remote operations of vehicles,presents a significant challenge in transmitting large volumes of image data to a central server.This requirement outpaces the capacity of t...Remote driving,an emergent technology enabling remote operations of vehicles,presents a significant challenge in transmitting large volumes of image data to a central server.This requirement outpaces the capacity of traditional communication methods.To tackle this,we propose a novel framework using semantic communications,through a region of interest semantic segmentation method,to reduce the communication costs by transmitting meaningful semantic information rather than bit-wise data.To solve the knowledge base inconsistencies inherent in semantic communications,we introduce a blockchain-based edge-assisted system for managing diverse and geographically varied semantic segmentation knowledge bases.This system not only ensures the security of data through the tamper-resistant nature of blockchain but also leverages edge computing for efficient management.Additionally,the implementation of blockchain sharding handles differentiated knowledge bases for various tasks,thus boosting overall blockchain efficiency.Experimental results show a great reduction in latency by sharding and an increase in model accuracy,confirming our framework's effectiveness.展开更多
文摘Weakly Supervised Semantic Segmentation(WSSS),which relies only on image-level labels,has attracted significant attention for its cost-effectiveness and scalability.Existing methods mainly enhance inter-class distinctions and employ data augmentation to mitigate semantic ambiguity and reduce spurious activations.However,they often neglect the complex contextual dependencies among image patches,resulting in incomplete local representations and limited segmentation accuracy.To address these issues,we propose the Context Patch Fusion with Class Token Enhancement(CPF-CTE)framework,which exploits contextual relations among patches to enrich feature repre-sentations and improve segmentation.At its core,the Contextual-Fusion Bidirectional Long Short-Term Memory(CF-BiLSTM)module captures spatial dependencies between patches and enables bidirectional information flow,yield-ing a more comprehensive understanding of spatial correlations.This strengthens feature learning and segmentation robustness.Moreover,we introduce learnable class tokens that dynamically encode and refine class-specific semantics,enhancing discriminative capability.By effectively integrating spatial and semantic cues,CPF-CTE produces richer and more accurate representations of image content.Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 validate that CPF-CTE consistently surpasses prior WSSS methods.
基金supported by the National Natural Science Foundation of China(22072048)the Guangdong Provincial Department of Science and Technology(2021A1515010128 and 2022A0505050013).
文摘Succinonitrile(SN)-based polymer plastic crystal electrolytes(PPCEs)are regarded as promising candidates for lithium metal batteries but suffer from serious side reactions with Li metal.Herein,we propose a multi-dimensional optimization strategy to alleviate the side reactions between SN and Li metal,and develop a highly stable poly-vinylethylene carbonate-based PPCE(PPCE-VEC).Moreover,we identify the intrinsic factors of multi-dimensional polymer structures on the electrolyte stability by three typical classes of polyesters.The PPCE-VEC constructed by in situ polymerization exhibits much better stability than poly-vinylene carbonate-based PPCE(PPCE-VCA)and poly-trifluoroethyl acrylate-based PPCE(PPCE-TFA),which is verified by its fewer SN-decomposition species in X-ray photoelectron spectroscopy(XPS)and outstanding full cell performance.The PPCE-VEC-enabled LiNi_(0.6)Co_(0.2)Mn_(0.2)O_(2)full cell achieve 73.7%capacity retention after 1400 cycles,which outperforms PPCE-VCA-and PPCE-TFA-enabled full cells(61.9%and 46.9%).Spectral analysis and theoretical calculation reveal that the high solvation ability of the carbonyl site,flexible polymer chain,and homogeneous electrolyte phase of PPCE-VEC are favorable to maximizing competition coordination with Li^(+)to weaken the Li^(+)–SN binding and shape an anion-rich solvation structure.This optimized polymer-involved Li^(+)solvation enhances SN stability and facilitates the formation of B/F enriched solid-electrolyte interphase(SEI),thus significantly improving PPCE stability.
基金supported in part by the National Natural Science Foundation of China under Grant 62262073in part by the Yunnan Provincial Ten Thousand People Program for Young Top Talents under Grant YNWR-QNBJ-2019-237in part by the Yunnan Provincial Major Science and Technology Special Program under Grant 202402AD080002.
文摘In the age of big data,ensuring data privacy while enabling efficient encrypted data retrieval has become a critical challenge.Traditional searchable encryption schemes face difficulties in handling complex semantic queries.Additionally,they typically rely on honest but curious cloud servers,which introduces the risk of repudiation.Furthermore,the combined operations of search and verification increase system load,thereby reducing performance.Traditional verification mechanisms,which rely on complex hash constructions,suffer from low verification efficiency.To address these challenges,this paper proposes a blockchain-based contextual semantic-aware ciphertext retrieval scheme with efficient verification.Building on existing single and multi-keyword search methods,the scheme uses vector models to semantically train the dataset,enabling it to retain semantic information and achieve context-aware encrypted retrieval,significantly improving search accuracy.Additionally,a blockchain-based updatable master-slave chain storage model is designed,where the master chain stores encrypted keyword indexes and the slave chain stores verification information generated by zero-knowledge proofs,thus balancing system load while improving search and verification efficiency.Finally,an improved non-interactive zero-knowledge proof mechanism is introduced,reducing the computational complexity of verification and ensuring efficient validation of search results.Experimental results demonstrate that the proposed scheme offers stronger security,balanced overhead,and higher search verification efficiency.
基金supported by the National Science and Technology Council of under Grant NSTC 114-2221-E-130-007.
文摘This paper presents an intelligent patrol and security robot integrating 2D LiDAR and RGB-D vision sensors to achieve semantic simultaneous localization and mapping(SLAM),real-time object recognition,and dynamic obstacle avoidance.The system employs the YOLOv7 deep-learning framework for semantic detection and SLAM for localization and mapping,fusing geometric and visual data to build a high-fidelity 2D semantic map.This map enables the robot to identify and project object information for improved situational awareness.Experimental results show that object recognition reached 95.4%mAP@0.5.Semantic completeness increased from 68.7%(single view)to 94.1%(multi-view)with an average position error of 3.1 cm.During navigation,the robot achieved 98.0%reliability,avoided moving obstacles in 90.0%of encounters,and replanned paths in 0.42 s on average.The integration of LiDAR-based SLAMwith deep-learning–driven semantic perception establishes a robust foundation for intelligent,adaptive,and safe robotic navigation in dynamic environments.
基金supported by the National Key Research and Development Program of China(2020AAA0109300)the Shanghai Collaborative Innovation Center of data intelligence technology(No.0232-A1-8900-24-13).
文摘Chinese abbreviations improve communicative efficiency by extracting key components from longer expressions.They are widely used in both daily communication and professional domains.However,existing abbreviation generation methods still face two major challenges.First,sequence-labeling-based approaches often neglect contextual meaning by making binary decisions at the character level,leading to abbreviations that fail to capture semantic completeness.Second,generation-basedmethods rely heavily on a single decoding process,which frequently produces correct abbreviations but ranks them lower due to inadequate semantic evaluation.To address these limitations,we propose a novel two-stage frameworkwithGeneration–Iterative Optimization forAbbreviation(GIOA).In the first stage,we design aChain-of-Thought prompting strategy and incorporate definitional and situational contexts to generate multiple abbreviation candidates.In the second stage,we introduce a Semantic Preservation Dynamic Adjustment mechanism that alternates between character-level importance estimation and semantic restoration to optimize candidate ranking.Experiments on two public benchmark datasets show that our method outperforms existing state-of-the-art approaches,achieving Hit@1 improvements of 15.15%and 13.01%,respectively,while maintaining consistent results in Hit@3.
基金provided by the Science Research Project of Hebei Education Department under grant No.BJK2024115.
文摘High-resolution remote sensing images(HRSIs)are now an essential data source for gathering surface information due to advancements in remote sensing data capture technologies.However,their significant scale changes and wealth of spatial details pose challenges for semantic segmentation.While convolutional neural networks(CNNs)excel at capturing local features,they are limited in modeling long-range dependencies.Conversely,transformers utilize multihead self-attention to integrate global context effectively,but this approach often incurs a high computational cost.This paper proposes a global-local multiscale context network(GLMCNet)to extract both global and local multiscale contextual information from HRSIs.A detail-enhanced filtering module(DEFM)is proposed at the end of the encoder to refine the encoder outputs further,thereby enhancing the key details extracted by the encoder and effectively suppressing redundant information.In addition,a global-local multiscale transformer block(GLMTB)is proposed in the decoding stage to enable the modeling of rich multiscale global and local information.We also design a stair fusion mechanism to transmit deep semantic information from deep to shallow layers progressively.Finally,we propose the semantic awareness enhancement module(SAEM),which further enhances the representation of multiscale semantic features through spatial attention and covariance channel attention.Extensive ablation analyses and comparative experiments were conducted to evaluate the performance of the proposed method.Specifically,our method achieved a mean Intersection over Union(mIoU)of 86.89%on the ISPRS Potsdam dataset and 84.34%on the ISPRS Vaihingen dataset,outperforming existing models such as ABCNet and BANet.
文摘Advanced traffic monitoring systems encounter substantial challenges in vehicle detection and classification due to the limitations of conventional methods,which often demand extensive computational resources and struggle with diverse data acquisition techniques.This research presents a novel approach for vehicle classification and recognition in aerial image sequences,integrating multiple advanced techniques to enhance detection accuracy.The proposed model begins with preprocessing using Multiscale Retinex(MSR)to enhance image quality,followed by Expectation-Maximization(EM)Segmentation for precise foreground object identification.Vehicle detection is performed using the state-of-the-art YOLOv10 framework,while feature extraction incorporates Maximally Stable Extremal Regions(MSER),Dense Scale-Invariant Feature Transform(Dense SIFT),and Zernike Moments Features to capture distinct object characteristics.Feature optimization is further refined through a Hybrid Swarm-based Optimization algorithm,ensuring optimal feature selection for improved classification performance.The final classification is conducted using a Vision Transformer,leveraging its robust learning capabilities for enhanced accuracy.Experimental evaluations on benchmark datasets,including UAVDT and the Unmanned Aerial Vehicle Intruder Dataset(UAVID),demonstrate the superiority of the proposed approach,achieving an accuracy of 94.40%on UAVDT and 93.57%on UAVID.The results highlight the efficacy of the model in significantly enhancing vehicle detection and classification in aerial imagery,outperforming existing methodologies and offering a statistically validated improvement for intelligent traffic monitoring systems compared to existing approaches.
文摘Weakly supervised semantic segmentation(WSSS)is a tricky task,which only provides category information for segmentation prediction.Thus,the key stage of WSSS is to generate the pseudo labels.For convolutional neural network(CNN)based methods,in which class activation mapping(CAM)is proposed to obtain the pseudo labels,and only concentrates on the most discriminative parts.Recently,transformer-based methods utilize attention map from the multi-headed self-attention(MHSA)module to predict pseudo labels,which usually contain obvious background noise and incoherent object area.To solve the above problems,we use the Conformer as our backbone,which is a parallel network based on convolutional neural network(CNN)and Transformer.The two branches generate pseudo labels and refine them independently,and can effectively combine the advantages of CNN and Transformer.However,the parallel structure is not close enough in the information communication.Thus,parallel structure can result in poor details about pseudo labels,and the background noise still exists.To alleviate this problem,we propose enhancing convolution CAM(ECCAM)model,which have three improved modules based on enhancing convolution,including deeper stem(DStem),convolutional feed-forward network(CFFN)and feature coupling unit with convolution(FCUConv).The ECCAM could make Conformer have tighter interaction between CNN and Transformer branches.After experimental verification,the improved modules we propose can help the network perceive more local information from images,making the final segmentation results more refined.Compared with similar architecture,our modules greatly improve the semantic segmentation performance and achieve70.2%mean intersection over union(mIoU)on the PASCAL VOC 2012 dataset.
基金supported by the Funds for Central-Guided Local Science and Technology Development(Grant No.202407AC110005)Key Technologies for the Construction of a Whole-Process Intelligent Service System for Neuroendocrine Neoplasm.Supported by 2023 Opening Research Fund of Yunnan Key Laboratory of Digital Communications(YNJTKFB-20230686,YNKLDC-KFKT-202304).
文摘In image analysis,high-precision semantic segmentation predominantly relies on supervised learning.Despite significant advancements driven by deep learning techniques,challenges such as class imbalance and dynamic performance evaluation persist.Traditional weighting methods,often based on pre-statistical class counting,tend to overemphasize certain classes while neglecting others,particularly rare sample categories.Approaches like focal loss and other rare-sample segmentation techniques introduce multiple hyperparameters that require manual tuning,leading to increased experimental costs due to their instability.This paper proposes a novel CAWASeg framework to address these limitations.Our approach leverages Grad-CAM technology to generate class activation maps,identifying key feature regions that the model focuses on during decision-making.We introduce a Comprehensive Segmentation Performance Score(CSPS)to dynamically evaluate model performance by converting these activation maps into pseudo mask and comparing them with Ground Truth.Additionally,we design two adaptive weights for each class:a Basic Weight(BW)and a Ratio Weight(RW),which the model adjusts during training based on real-time feedback.Extensive experiments on the COCO-Stuff,CityScapes,and ADE20k datasets demonstrate that our CAWASeg framework significantly improves segmentation performance for rare sample categories while enhancing overall segmentation accuracy.The proposed method offers a robust and efficient solution for addressing class imbalance in semantic segmentation tasks.
基金National Natural Science Foundation of China(Nos.42301473,42271424,42171397)Chinese Postdoctoral Innovation Talents Support Program(No.BX20230299)+2 种基金China Postdoctoral Science Foundation(No.2023M742884)Natural Science Foundation of Sichuan Province(Nos.24NSFSC2264,2025ZNSFSC0322)Key Research and Development Project of Sichuan Province(No.24ZDYF0633).
文摘As a key node of modern transportation network,the informationization management of road tunnels is crucial to ensure the operation safety and traffic efficiency.However,the existing tunnel vehicle modeling methods generally have problems such as insufficient 3D scene description capability and low dynamic update efficiency,which are difficult to meet the demand of real-time accurate management.For this reason,this paper proposes a vehicle twin modeling method for road tunnels.This approach starts from the actual management needs,and supports multi-level dynamic modeling from vehicle type,size to color by constructing a vehicle model library that can be flexibly invoked;at the same time,semantic constraint rules with geometric layout,behavioral attributes,and spatial relationships are designed to ensure that the virtual model matches with the real model with a high degree of similarity;ultimately,the prototype system is constructed and the case region is selected for the case study,and the dynamic vehicle status in the tunnel is realized by integrating real-time monitoring data with semantic constraints for precise virtual-real mapping.Finally,the prototype system is constructed and case experiments are conducted in selected case areas,which are combined with real-time monitoring data to realize dynamic updating and three-dimensional visualization of vehicle states in tunnels.The experiments show that the proposed method can run smoothly with an average rendering efficiency of 17.70 ms while guaranteeing the modeling accuracy(composite similarity of 0.867),which significantly improves the real-time and intuitive tunnel management.The research results provide reliable technical support for intelligent operation and emergency response of road tunnels,and offer new ideas for digital twin modeling of complex scenes.
基金co-supported by the National Key R&D Program of China(No.2023YFB4704400)the Zhejiang Provincial Natural Science Foundation of China(No.LQ24F030012)the National Natural Science Foundation of China General Project(No.62373033)。
文摘A composite anti-disturbance predictive control strategy employing a Multi-dimensional Taylor Network(MTN)is presented for unmanned systems subject to time-delay and multi-source disturbances.First,the multi-source disturbances are addressed according to their specific characteristics as follows:(A)an MTN data-driven model,which is used for uncertainty description,is designed accompanied with the mechanism model to represent the unmanned systems;(B)an adaptive MTN filter is used to remove the influence of the internal disturbance;(C)an MTN disturbance observer is constructed to estimate and compensate for the influence of the external disturbance;(D)the Extended Kalman Filter(EKF)algorithm is utilized as the learning mechanism for MTNs.Second,to address the time-delay effect,a recursiveτstep-ahead MTN predictive model is designed utilizing recursive technology,aiming to mitigate the impact of time-delay,and the EKF algorithm is employed as its learning mechanism.Then,the MTN predictive control law is designed based on the quadratic performance index.By implementing the proposed composite controller to unmanned systems,simultaneous feedforward compensation and feedback suppression to the multi-source disturbances are conducted.Finally,the convergence of the MTN and the stability of the closed-loop system are established utilizing the Lyapunov theorem.Two exemplary applications of unmanned systems involving unmanned vehicle and rigid spacecraft are presented to validate the effectiveness of the proposed approach.
基金funded by the Natural Science Foundation of the Jiangsu Higher Education Institutions of China(grant number 22KJD440001)Changzhou Science&Technology Program(grant number CJ20220232).
文摘The Internet of Vehicles (IoV) has become an important direction in the field of intelligent transportation, in which vehicle positioning is a crucial part. SLAM (Simultaneous Localization and Mapping) technology plays a crucial role in vehicle localization and navigation. Traditional Simultaneous Localization and Mapping (SLAM) systems are designed for use in static environments, and they can result in poor performance in terms of accuracy and robustness when used in dynamic environments where objects are in constant movement. To address this issue, a new real-time visual SLAM system called MG-SLAM has been developed. Based on ORB-SLAM2, MG-SLAM incorporates a dynamic target detection process that enables the detection of both known and unknown moving objects. In this process, a separate semantic segmentation thread is required to segment dynamic target instances, and the Mask R-CNN algorithm is applied on the Graphics Processing Unit (GPU) to accelerate segmentation. To reduce computational cost, only key frames are segmented to identify known dynamic objects. Additionally, a multi-view geometry method is adopted to detect unknown moving objects. The results demonstrate that MG-SLAM achieves higher precision, with an improvement from 0.2730 m to 0.0135 m in precision. Moreover, the processing time required by MG-SLAM is significantly reduced compared to other dynamic scene SLAM algorithms, which illustrates its efficacy in locating objects in dynamic scenes.
基金supported by the project“GEF9874:Strengthening Coordinated Approaches to Reduce Invasive Alien Species(lAS)Threats to Globally Significant Agrobiodiversity and Agroecosystems in China”funding from the Excellent Talent Training Funding Project in Dongcheng District,Beijing,with project number 2024-dchrcpyzz-9.
文摘Ecological monitoring vehicles are equipped with a range of sensors and monitoring devices designed to gather data on ecological and environmental factors.These vehicles are crucial in various fields,including environmental science research,ecological and environmental monitoring projects,disaster response,and emergency management.A key method employed in these vehicles for achieving high-precision positioning is LiDAR(lightlaser detection and ranging)-Visual Simultaneous Localization and Mapping(SLAM).However,maintaining highprecision localization in complex scenarios,such as degraded environments or when dynamic objects are present,remains a significant challenge.To address this issue,we integrate both semantic and texture information from LiDAR and cameras to enhance the robustness and efficiency of data registration.Specifically,semantic information simplifies the modeling of scene elements,reducing the reliance on dense point clouds,which can be less efficient.Meanwhile,visual texture information complements LiDAR-Visual localization by providing additional contextual details.By incorporating semantic and texture details frompaired images and point clouds,we significantly improve the quality of data association,thereby increasing the success rate of localization.This approach not only enhances the operational capabilities of ecological monitoring vehicles in complex environments but also contributes to improving the overall efficiency and effectiveness of ecological monitoring and environmental protection efforts.
文摘Lower back pain is one of the most common medical problems in the world and it is experienced by a huge percentage of people everywhere.Due to its ability to produce a detailed view of the soft tissues,including the spinal cord,nerves,intervertebral discs,and vertebrae,Magnetic Resonance Imaging is thought to be the most effective method for imaging the spine.The semantic segmentation of vertebrae plays a major role in the diagnostic process of lumbar diseases.It is difficult to semantically partition the vertebrae in Magnetic Resonance Images from the surrounding variety of tissues,including muscles,ligaments,and intervertebral discs.U-Net is a powerful deep-learning architecture to handle the challenges of medical image analysis tasks and achieves high segmentation accuracy.This work proposes a modified U-Net architecture namely MU-Net,consisting of the Meijering convolutional layer that incorporates the Meijering filter to perform the semantic segmentation of lumbar vertebrae L1 to L5 and sacral vertebra S1.Pseudo-colour mask images were generated and used as ground truth for training the model.The work has been carried out on 1312 images expanded from T1-weighted mid-sagittal MRI images of 515 patients in the Lumbar Spine MRI Dataset publicly available from Mendeley Data.The proposed MU-Net model for the semantic segmentation of the lumbar vertebrae gives better performance with 98.79%of pixel accuracy(PA),98.66%of dice similarity coefficient(DSC),97.36%of Jaccard coefficient,and 92.55%mean Intersection over Union(mean IoU)metrics using the mentioned dataset.
基金supported by the HFIPS Director’s Foundation(YZJJ202207-TS),the National Natural Science Foundation of China(82371931)the Natural Science Foundation of Anhui Province(2008085MC69)+3 种基金the Natural Science Foundation of Hefei City(2021033)the General Scientific Research Project of Anhui Provincial Health Commission(AHWJ2021b150)the Collaborative Innovation Program of Hefei Science Center,CAS(2021HSC-CIP013)the Anhui Province Key Research and Development Project(202204295107020004).
文摘Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is crucial for computationally limited portable devices such as augmented reality and virtual reality.With the rapid advancements in deep learning,many network models have been developed specifically for eye image segmentation.Some methods divide the segmentation process into multiple stages to achieve model parameter miniaturization while enhancing output through post processing techniques to improve segmentation accuracy.These approaches significantly increase the inference time.Other networks adopt more complex encoding and decoding modules to achieve end-to-end output,which requires substantial computation.Therefore,balancing the model’s size,accuracy,and computational complexity is essential.To address these challenges,we propose a lightweight asymmetric UNet architecture and a projection loss function.We utilize ResNet-3 layer blocks to enhance feature extraction efficiency in the encoding stage.In the decoding stage,we employ regular convolutions and skip connections to upscale the feature maps from the latent space to the original image size,balancing the model size and segmentation accuracy.In addition,we leverage the geometric features of the eye region and design a projection loss function to further improve the segmentation accuracy without adding any additional inference computational cost.We validate our approach on the OpenEDS2019 dataset for virtual reality and achieve state-of-the-art performance with 95.33%mean intersection over union(mIoU).Our model has only 0.63M parameters and 350 FPS,which are 68%and 200%of the state-of-the-art model RITNet,respectively.
文摘Semantic segmentation is a core task in computer vision that allows AI models to interact and understand their surrounding environment. Similarly to how humans subconsciously segment scenes, this ability is crucial for scene understanding. However, a challenge many semantic learning models face is the lack of data. Existing video datasets are limited to short, low-resolution videos that are not representative of real-world examples. Thus, one of our key contributions is a customized semantic segmentation version of the Walking Tours Dataset that features hour-long, high-resolution, real-world data from tours of different cities. Additionally, we evaluate the performance of open-vocabulary, semantic model OpenSeeD on our own custom dataset and discuss future implications.
文摘Few-shot image semantic segmentation aims to achieve pixel-level classification for novel classes using only a few labeled examples.The method first trains the segmentation model on base classes,and then adapts it to novel classes.Although existing methods have achieved remarkable performance in few-shot image semantic segmentation,they still face the following challenges.Traditional methods typically rely on mask average pooling to generate single-category prototype vectors and perform feature matching via metric learning,but they exhibit significant limitations in modeling inter-category relationships and addressing complex background interference.Inspired by the analogy-based transfer mechanisms in cognitive psychology,we propose a Generalized Prototype Network(GPNet)to enhance the model's generalization ability for unseen categories and improve robustness in feature matching.GPNet consists of two key modules.The first is a generalized prototype enhancement module,which explores potential inter-category relationships to construct more discriminative category prototype representations.The second is a multi-scale feature alignment module,which dynamically aligns support and query features across multiple scales using an attention mechanism,thus mitigating background interference in complex scenarios.Experimental results demonstrate that the proposed method significantly outperforms existing state-of-the-art approaches on several few-shot semantic segmentation benchmarks,validating its effectiveness and generalization capabilities.
基金supported in part by the National Key Research and Development Program of China under Grant 2021YFA1000500(4)in part by the Natural Science Foundation of China(NSFC)under Grant 62293484,Grant U22B2001,Grant 62425110,Grant 62227801,Grant 62442106.
文摘In recent years,deep learning-based semantic communications have shown great potential to enhance the performance of communication systems.This has led to the belief that semantic communications represent a breakthrough beyond the Shannon paradigm and will play an essential role in future communications.To narrow the gap between current research and future vision,after an overview of semantic communications,this article presents and discusses ten fundamental and critical challenges in today’s semantic communication field.These challenges are divided into theory foundation,system design,and practical implementation.Challenges related to the theory foundation including semantic capacity,entropy,and rate-distortion are discussed first.Then,the system design challenges encompassing architecture,knowledge base,joint semantic-channel coding,tailored transmission scheme,and impairment are posed.The last two challenges associated with the practical implementation lie in cross-layer optimization for networks and standardization.For each challenge,efforts to date and thoughtful insights are provided.
基金supported in part by the Innovation and Entrepreneurship Training Program for Chinese College Students(No.202410128019)in part by JST ASPIRE Grant Number JPMJAP2325in part by Support Center for Advanced Telecommunications Technology Research(SCAT).
文摘With the rapid development of artificial intelligence and the Internet of Things,along with the growing demand for privacy-preserving transmission,the need for efficient and secure communication systems has become increasingly urgent.Traditional communication methods transmit data at the bit level without considering its semantic significance,leading to redundant transmission overhead and reduced efficiency.Semantic communication addresses this issue by extracting and transmitting only the mostmeaningful semantic information,thereby improving bandwidth efficiency.However,despite reducing the volume of data,it remains vulnerable to privacy risks,as semantic features may still expose sensitive information.To address this,we propose an entropy-bottleneck-based privacy protection mechanism for semantic communication.Our approach uses semantic segmentation to partition images into regions of interest(ROI)and regions of non-interest(RONI)based on the receiver’s needs,enabling differentiated semantic transmission.By focusing transmission on ROIs,bandwidth usage is optimized,and non-essential data is minimized.The entropy bottleneck model probabilistically encodes the semantic information into a compact bit stream,reducing correlation between the transmitted content and the original data,thus enhancing privacy protection.The proposed framework is systematically evaluated in terms of compression efficiency,semantic fidelity,and privacy preservation.Through comparative experiments with traditional and state-of-the-art methods,we demonstrate that the approach significantly reduces data transmission,maintains the quality of semantically important regions,and ensures robust privacy protection.
基金supported in part by the National Natural Science Foundation of China under Grant No.62062031in part by the MIC/SCOPE#JP235006102+2 种基金in part by JST ASPIRE Grant Number JPMJAP2325in part by ROIS NII Open Collaborative Research under Grant 24S0601in part by collaborative research with Toyota Motor Corporation,Japan。
文摘Remote driving,an emergent technology enabling remote operations of vehicles,presents a significant challenge in transmitting large volumes of image data to a central server.This requirement outpaces the capacity of traditional communication methods.To tackle this,we propose a novel framework using semantic communications,through a region of interest semantic segmentation method,to reduce the communication costs by transmitting meaningful semantic information rather than bit-wise data.To solve the knowledge base inconsistencies inherent in semantic communications,we introduce a blockchain-based edge-assisted system for managing diverse and geographically varied semantic segmentation knowledge bases.This system not only ensures the security of data through the tamper-resistant nature of blockchain but also leverages edge computing for efficient management.Additionally,the implementation of blockchain sharding handles differentiated knowledge bases for various tasks,thus boosting overall blockchain efficiency.Experimental results show a great reduction in latency by sharding and an increase in model accuracy,confirming our framework's effectiveness.
