Pub Date : 2024-10-14DOI: 10.1109/LSP.2024.3479918
T. Averty;A. O. Boudraa;D. Daré-Emzivat
Most natural matrices that incorporate information about a graph are the adjacency and the Laplacian matrices. These algebraic representations govern the fundamental concepts and tools in graph signal processing even though they reveal information in different ways. Furthermore, in the context of spectral graph classification, the problem of cospectrality may arise and it is not well handled by these matrices. Thus, the question of finding the best graph representation matrix still stands. In this letter, a new family of representations that well captures information about graphs and also allows to find the standard representation matrices, is introduced. This family of unified matrices well captures the graph information and extends the recent works of the literature. Two properties are proven, namely its positive semidefiniteness and the monotonicity of their eigenvalues. Reported experimental results of spectral graph classification highlight the potential and the added value of this new family of matrices, and evidence that the best representation depends upon the structure of the underlying graph.
{"title":"A New Family of Graph Representation Matrices: Application to Graph and Signal Classification","authors":"T. Averty;A. O. Boudraa;D. Daré-Emzivat","doi":"10.1109/LSP.2024.3479918","DOIUrl":"https://doi.org/10.1109/LSP.2024.3479918","url":null,"abstract":"Most natural matrices that incorporate information about a graph are the adjacency and the Laplacian matrices. These algebraic representations govern the fundamental concepts and tools in graph signal processing even though they reveal information in different ways. Furthermore, in the context of spectral graph classification, the problem of cospectrality may arise and it is not well handled by these matrices. Thus, the question of finding the best graph representation matrix still stands. In this letter, a new family of representations that well captures information about graphs and also allows to find the standard representation matrices, is introduced. This family of unified matrices well captures the graph information and extends the recent works of the literature. Two properties are proven, namely its positive semidefiniteness and the monotonicity of their eigenvalues. Reported experimental results of spectral graph classification highlight the potential and the added value of this new family of matrices, and evidence that the best representation depends upon the structure of the underlying graph.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2935-2939"},"PeriodicalIF":3.2,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142524186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-14DOI: 10.1109/LSP.2024.3479934
Penghui Lai;Yaru Shan;Fanggang Wang;Shilian Wang;Peiguo Liu
The orthogonal time frequency space (OTFS) modulation is resilient to the Doppler effect and thus is employed in high-mobility communications. In earlier work, the equivalent representations of the OTFS transmission were established, revealing the profound impact of fractional delay on channel sparsity. However, these representations tend to be distorted and cumbersome when accounting for digital reception and non-ideal filtering. In this letter, we propose an alternative representation with low distortion that uses the variable-fractional-delay filter to characterize the fractional delays in a conventional digital transceiver. In our proposed method, the approximation of the received signal is improved, and the error performance is enhanced compared to the original approaches. At last, the simulations show that our proposed representation is valid.
{"title":"Approximation of Non-Ideal Filtering in OTFS via Variable Fractional Delay","authors":"Penghui Lai;Yaru Shan;Fanggang Wang;Shilian Wang;Peiguo Liu","doi":"10.1109/LSP.2024.3479934","DOIUrl":"https://doi.org/10.1109/LSP.2024.3479934","url":null,"abstract":"The orthogonal time frequency space (OTFS) modulation is resilient to the Doppler effect and thus is employed in high-mobility communications. In earlier work, the equivalent representations of the OTFS transmission were established, revealing the profound impact of fractional delay on channel sparsity. However, these representations tend to be distorted and cumbersome when accounting for digital reception and non-ideal filtering. In this letter, we propose an alternative representation with low distortion that uses the variable-fractional-delay filter to characterize the fractional delays in a conventional digital transceiver. In our proposed method, the approximation of the received signal is improved, and the error performance is enhanced compared to the original approaches. At last, the simulations show that our proposed representation is valid.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2990-2994"},"PeriodicalIF":3.2,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142573601","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-10DOI: 10.1109/LSP.2024.3478211
Yiya Hao;Feifei Xiong;Bei Li;Nai Ding;Jinwei Feng
We present a neural speech quality assessment model with speaker embedding. This model, i.e., EMDSQA, can precisely predict the Mean Opinion Score (MOS) of speech quality during online communications. Intrusive speech quality assessment methods such as perceptual objective listening quality analysis (POLQA) are not practical for online communications because every piece of degraded speech requires a corresponding clean reference. Non-intrusive methods can assess the quality of online speech, but have not reached the accuracy and robustness required for real-world applications. EMDSQA extracts the speaker embedding using an independent pipeline and feeds it as a prior feature to a self-attention-based MOS prediction model. Since EMDSQA does not need the corresponding clean reference, it is practical for real-world communication applications. An open-source test corpus, featuring real-world data, was also developed. Experimental results show that EMDSQA achieves a 0.92 Pearson correlation coefficient with the MOS measured from humans, surpassing other state-of-the-art intrusive or non-intrusive methods.
我们提出了一种具有说话者嵌入功能的神经语音质量评估模型。该模型（即 EMDSQA）可精确预测在线通信中语音质量的平均意见分（MOS）。感知客观听力质量分析（POLQA）等侵入式语音质量评估方法在在线通信中并不实用，因为每一段降级语音都需要相应的干净参考。非侵入式方法可以评估在线语音质量，但尚未达到实际应用所需的准确性和鲁棒性。EMDSQA 使用独立管道提取说话者嵌入，并将其作为先验特征输入基于自我关注的 MOS 预测模型。由于 EMDSQA 不需要相应的干净参考，因此在实际通信应用中非常实用。此外，还开发了一个以真实世界数据为特色的开源测试语料库。实验结果表明，EMDSQA 与人工测量的 MOS 之间的皮尔逊相关系数达到了 0.92，超过了其他最先进的侵入式或非侵入式方法。
{"title":"EMDSQA: A Neural Speech Quality Assessment Model With Speaker Embedding","authors":"Yiya Hao;Feifei Xiong;Bei Li;Nai Ding;Jinwei Feng","doi":"10.1109/LSP.2024.3478211","DOIUrl":"https://doi.org/10.1109/LSP.2024.3478211","url":null,"abstract":"We present a neural speech quality assessment model with speaker embedding. This model, i.e., EMDSQA, can precisely predict the Mean Opinion Score (MOS) of speech quality during online communications. Intrusive speech quality assessment methods such as perceptual objective listening quality analysis (POLQA) are not practical for online communications because every piece of degraded speech requires a corresponding clean reference. Non-intrusive methods can assess the quality of online speech, but have not reached the accuracy and robustness required for real-world applications. EMDSQA extracts the speaker embedding using an independent pipeline and feeds it as a prior feature to a self-attention-based MOS prediction model. Since EMDSQA does not need the corresponding clean reference, it is practical for real-world communication applications. An open-source test corpus, featuring real-world data, was also developed. Experimental results show that EMDSQA achieves a 0.92 Pearson correlation coefficient with the MOS measured from humans, surpassing other state-of-the-art intrusive or non-intrusive methods.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"3064-3068"},"PeriodicalIF":3.2,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10713506","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142600134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-10DOI: 10.1109/LSP.2024.3478109
Fan Wang;Zhangjie Fu;Xiang Zhang;Junjie Lu
Adversarial embedding for image steganography is a novel technology to effectively enhance the steganographic security of the traditional steganographic algorithms. However, the existing schemes still have room for further improvement in the design of optimization strategy and the steganographic post-processing of optimization failure. In this paper, we design the progressive probability optimizing strategy (PPO). It dynamically selects more efficient gradients to guide the optimization of the probability optimization in a progressive manner. Moreover, we propose a discarded stego recycling mechanism (DSR) to re-select the stego from the discarded stego set that have failed to deceive the target steganalyzer after the optimzation fails. In such way, the statistical distribution of the stego can still further approximate the cover, thus further improving the steganographic security on re-trained steganalyzers in adversary-aware scenario. Comprehensive experiments show that compared with the existing advanced schemes, the proposed method boosts the security improvement against both the re-trained hand-crafted feature-based and deep leanring-based steganalysis models.
{"title":"Adversarial Embedding Steganography via Progressive Probability Optimizing and Discarded Stego Recycling","authors":"Fan Wang;Zhangjie Fu;Xiang Zhang;Junjie Lu","doi":"10.1109/LSP.2024.3478109","DOIUrl":"https://doi.org/10.1109/LSP.2024.3478109","url":null,"abstract":"Adversarial embedding for image steganography is a novel technology to effectively enhance the steganographic security of the traditional steganographic algorithms. However, the existing schemes still have room for further improvement in the design of optimization strategy and the steganographic post-processing of optimization failure. In this paper, we design the progressive probability optimizing strategy (PPO). It dynamically selects more efficient gradients to guide the optimization of the probability optimization in a progressive manner. Moreover, we propose a discarded stego recycling mechanism (DSR) to re-select the stego from the discarded stego set that have failed to deceive the target steganalyzer after the optimzation fails. In such way, the statistical distribution of the stego can still further approximate the cover, thus further improving the steganographic security on re-trained steganalyzers in adversary-aware scenario. Comprehensive experiments show that compared with the existing advanced schemes, the proposed method boosts the security improvement against both the re-trained hand-crafted feature-based and deep leanring-based steganalysis models.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2920-2924"},"PeriodicalIF":3.2,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142524170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Within the domain of multimodal communication, the compression of audio, image, and video information is well-established, but compressing haptic signals, including vibrotactile signals, remains challenging. Particularly with the enhancement of haptic signal sampling rate and degrees of freedom, there is a substantial increase in data volume. While existing algorithms have made progress in vibrotactile codecs, there remains significant room for improvement in compression ratios. We propose an innovative Nbeats Network-based Vibrotactile Codec (NNVC) that leverages the statistical characteristics of vibrotactile data. This advanced codec integrates the Nbeats network for precise vibrotactile prediction, residual quantization, efficient Run-Length Encoding, and Huffman coding. The algorithm not only captures the intricate details of vibrotactile signals but also ensures high-efficiency data compression. It exhibits robust overall performance in terms of Signal-to-Noise Ratio (SNR) and Peak Signal-to-Noise Ratio (PSNR), significantly surpassing the state-of-the-art.
{"title":"Efficient Vibrotactile Codec Based on Nbeats Network","authors":"Yiwen Xu;Dongfang Chen;Ying Fang;Yang Lu;Tiesong Zhao","doi":"10.1109/LSP.2024.3477251","DOIUrl":"https://doi.org/10.1109/LSP.2024.3477251","url":null,"abstract":"Within the domain of multimodal communication, the compression of audio, image, and video information is well-established, but compressing haptic signals, including vibrotactile signals, remains challenging. Particularly with the enhancement of haptic signal sampling rate and degrees of freedom, there is a substantial increase in data volume. While existing algorithms have made progress in vibrotactile codecs, there remains significant room for improvement in compression ratios. We propose an innovative Nbeats Network-based Vibrotactile Codec (NNVC) that leverages the statistical characteristics of vibrotactile data. This advanced codec integrates the Nbeats network for precise vibrotactile prediction, residual quantization, efficient Run-Length Encoding, and Huffman coding. The algorithm not only captures the intricate details of vibrotactile signals but also ensures high-efficiency data compression. It exhibits robust overall performance in terms of Signal-to-Noise Ratio (SNR) and Peak Signal-to-Noise Ratio (PSNR), significantly surpassing the state-of-the-art.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2845-2849"},"PeriodicalIF":3.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142447070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-09DOI: 10.1109/LSP.2024.3477254
Zhuo Chen;Xiaoming Niu;Jian Ding;Hong Wu;Zhiyang Liu
The orthogonal time frequency space (OTFS) has emerged as a promising modulation waveform for high-mobility wireless communications owing to its robust advantages of resisting Doppler effects. However, due to the limit of the frame duration, the fractional Doppler shift appears, which is a challenge for channel estimation in OTFS systems. In this letter, we formulate the channel estimation problem as a block sparse signal recovery issue and propose an adaptive pattern-coupled sparse Bayesian learning (APCSBL) method. To be specific, we introduce a pattern-coupled hierarchical Gaussian prior model to characterize the dependencies among adjacent channel coefficients. On this basis, an adaptive hyperparameter strategy is presented, in which we appropriately utilize various coupling parameters further to characterize the strength of the correlation between adjacent elements. Then we exploit the expectation maximization (EM) algorithm to update the hidden variables and the channel vector. Simulation results demonstrate that the proposed algorithm outperforms existing methods and works for various environments.
{"title":"Adaptive Pattern-Coupled Sparse Bayesian Learning for Channel Estimation in OTFS Systems","authors":"Zhuo Chen;Xiaoming Niu;Jian Ding;Hong Wu;Zhiyang Liu","doi":"10.1109/LSP.2024.3477254","DOIUrl":"https://doi.org/10.1109/LSP.2024.3477254","url":null,"abstract":"The orthogonal time frequency space (OTFS) has emerged as a promising modulation waveform for high-mobility wireless communications owing to its robust advantages of resisting Doppler effects. However, due to the limit of the frame duration, the fractional Doppler shift appears, which is a challenge for channel estimation in OTFS systems. In this letter, we formulate the channel estimation problem as a block sparse signal recovery issue and propose an adaptive pattern-coupled sparse Bayesian learning (APCSBL) method. To be specific, we introduce a pattern-coupled hierarchical Gaussian prior model to characterize the dependencies among adjacent channel coefficients. On this basis, an adaptive hyperparameter strategy is presented, in which we appropriately utilize various coupling parameters further to characterize the strength of the correlation between adjacent elements. Then we exploit the expectation maximization (EM) algorithm to update the hidden variables and the channel vector. Simulation results demonstrate that the proposed algorithm outperforms existing methods and works for various environments.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2895-2899"},"PeriodicalIF":3.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142524115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-09DOI: 10.1109/LSP.2024.3477298
Haoqian Wang;Zhongyang Xing;Zhongjie Xu;Xiangai Cheng;Teng Li
In this study, we explore poor edge reconstruction in image super-resolution (SR) tasks, emphasizing the significance of enhancing edge details identified through visual analysis. Existing SR networks typically optimize their network architectures, enabling complete feature extraction from feature maps. This is because the management of spatial and channel information during SR is often pivotal to the network's feature extraction capacity. Despite continuous improvements, directly comparing SR and high-resolution (HR) images through differential mapping reveals the suboptimal performance of these methods in edge reconstruction. In this paper, we introduce a edgey-aware attention transformer (EAT), which focuses on edge reconstruction while maintaining the effective original low frequency information retrieval. Our framework utilizes deformable convolution (DC) to adaptively extract edge features. Then feature enhancement techniques are employed to intensify edge-sensitive features. Furthermore, extensive experiments demonstrate our EAT's exceptional quantitative and visual results, which surpass most benchmarks. This validates the EAT's effectiveness when compared to state-of-the-art models. The code is available at https://github.com/ImWangHaoqian/EAT .
在这项研究中，我们探讨了图像超分辨率（SR）任务中的边缘重建问题，强调了通过视觉分析增强边缘细节的重要性。现有的 SR 网络通常会优化其网络架构，以便从特征图中完整提取特征。这是因为在 SR 过程中，空间和通道信息的管理往往对网络的特征提取能力至关重要。尽管不断改进，但通过差分映射直接比较 SR 和高分辨率（HR）图像发现，这些方法在边缘重建方面的性能并不理想。在本文中，我们介绍了一种边缘感知注意力转换器（EAT），它侧重于边缘重建，同时保持有效的原始低频信息检索。我们的框架利用可变形卷积（DC）自适应地提取边缘特征。然后采用特征增强技术来强化边缘敏感特征。此外，大量实验证明，我们的 EAT 在数量和视觉效果上都非常出色，超越了大多数基准测试。与最先进的模型相比，这验证了 EAT 的有效性。代码见 https://github.com/ImWangHaoqian/EAT。
{"title":"Edge-Aware Attention Transformer for Image Super-Resolution","authors":"Haoqian Wang;Zhongyang Xing;Zhongjie Xu;Xiangai Cheng;Teng Li","doi":"10.1109/LSP.2024.3477298","DOIUrl":"https://doi.org/10.1109/LSP.2024.3477298","url":null,"abstract":"In this study, we explore poor edge reconstruction in image super-resolution (SR) tasks, emphasizing the significance of enhancing edge details identified through visual analysis. Existing SR networks typically optimize their network architectures, enabling complete feature extraction from feature maps. This is because the management of spatial and channel information during SR is often pivotal to the network's feature extraction capacity. Despite continuous improvements, directly comparing SR and high-resolution (HR) images through differential mapping reveals the suboptimal performance of these methods in edge reconstruction. In this paper, we introduce a edgey-aware attention transformer (EAT), which focuses on edge reconstruction while maintaining the effective original low frequency information retrieval. Our framework utilizes deformable convolution (DC) to adaptively extract edge features. Then feature enhancement techniques are employed to intensify edge-sensitive features. Furthermore, extensive experiments demonstrate our EAT's exceptional quantitative and visual results, which surpass most benchmarks. This validates the EAT's effectiveness when compared to state-of-the-art models. The code is available at \u0000<uri>https://github.com/ImWangHaoqian/EAT</uri>\u0000.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2905-2909"},"PeriodicalIF":3.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142525779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-09DOI: 10.1109/LSP.2024.3478112
Xiaolong Yang;Bingrui Zhang;Mu Zhou;Ming Gao
In this letter, we propose an Integrated Sensing and Communications (ISAC) waveform design method based on the fusion of Orthogonal Time Frequency Space (OTFS) and Faster-Than-Nyquist (FTN). The objective is to maximize the communication data transmission rate while minimizing the sensing performance impact on the target parameter estimation. We first map the FTN symbols to OTFS waveform time domain for realizing symbol spacing compression and transmit them in time-varying channels. Then, an equalizer based on the Minimum Mean Square Error (MMSE) algorithm is used to eliminate the interference generated by the FTN. Simulation results show taking into account the system bit-error rate, the proposed method achieves an increase in the throughput as well as an improvement in the distance and velocity estimation of the target compared to the existing methods.
{"title":"Integrated Sensing and Communications Waveform Design for OTFS and FTN Fusion","authors":"Xiaolong Yang;Bingrui Zhang;Mu Zhou;Ming Gao","doi":"10.1109/LSP.2024.3478112","DOIUrl":"https://doi.org/10.1109/LSP.2024.3478112","url":null,"abstract":"In this letter, we propose an Integrated Sensing and Communications (ISAC) waveform design method based on the fusion of Orthogonal Time Frequency Space (OTFS) and Faster-Than-Nyquist (FTN). The objective is to maximize the communication data transmission rate while minimizing the sensing performance impact on the target parameter estimation. We first map the FTN symbols to OTFS waveform time domain for realizing symbol spacing compression and transmit them in time-varying channels. Then, an equalizer based on the Minimum Mean Square Error (MMSE) algorithm is used to eliminate the interference generated by the FTN. Simulation results show taking into account the system bit-error rate, the proposed method achieves an increase in the throughput as well as an improvement in the distance and velocity estimation of the target compared to the existing methods.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2870-2874"},"PeriodicalIF":3.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142452740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-09DOI: 10.1109/LSP.2024.3477263
Jae-Ho Choi;Kyung-Tae Kim
With the advent of deep learning (DL) for signal processing, the deployment of DL for radar-based crowd counting has yielded significant performance enhancement. Despite these advancements, current methodologies predominantly undergo validation in controlled conditions with limited subject movement variability, posing a challenge for practical usage. Addressing this gap, this letter first attempts the application of radar-based crowd counting in an unregulated and dense setting, capturing the radar reflections of up to 31 subjects in real-world scenarios, such as queues at restaurant kiosks. Furthermore, to address the complexities of such a challenging condition, we introduce a novel radar crowd counting model that utilizes a spatiotemporal transformer. The expremental results demonstrate the potentiality of the proposed model as a robust crowd counting system under the full realistic scenarios, as well as establish its superiority over the conventional radar-based crowd counting models.
{"title":"Radar-Based Crowd Counting in Real-World Environments With Spatiotemporal Transformer","authors":"Jae-Ho Choi;Kyung-Tae Kim","doi":"10.1109/LSP.2024.3477263","DOIUrl":"https://doi.org/10.1109/LSP.2024.3477263","url":null,"abstract":"With the advent of deep learning (DL) for signal processing, the deployment of DL for radar-based crowd counting has yielded significant performance enhancement. Despite these advancements, current methodologies predominantly undergo validation in controlled conditions with limited subject movement variability, posing a challenge for practical usage. Addressing this gap, this letter first attempts the application of radar-based crowd counting in an unregulated and dense setting, capturing the radar reflections of up to 31 subjects in real-world scenarios, such as queues at restaurant kiosks. Furthermore, to address the complexities of such a challenging condition, we introduce a novel radar crowd counting model that utilizes a spatiotemporal transformer. The expremental results demonstrate the potentiality of the proposed model as a robust crowd counting system under the full realistic scenarios, as well as establish its superiority over the conventional radar-based crowd counting models.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2900-2904"},"PeriodicalIF":3.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142524171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-08DOI: 10.1109/LSP.2024.3476208
Zhiyu Jiang;Ye Yuan;Yuan Yuan
Few-shot semantic segmentation (FSS) is crucial for image interpretation, yet it is constrained by requirements for extensive base data and a narrow focus on foreground-background differentiation. This work introduces Data-free Few-shot Semantic Segmentation (DFSS), a task that requires limited labeled images and forgoes the need for extensive base data, allowing for comprehensive image segmentation. The proposed method utilizes the Segment Anything Model (SAM) for its generalization capabilities. The Prototypical Metric Segment Anything Model is introduced, featuring an initial segmentation phase followed by prototype matching, effectively addressing the learning challenges posed by limited data. To enhance discrimination in multi-class segmentation, the Supervised Prototypical Contrastive Loss (SPCL) is designed to refine prototype features, ensuring intra-class cohesion and inter-class separation. To further accommodate intra-class variability, the Adaptive Prototype Update (APU) strategy dynamically refines prototypes, adapting the model to class heterogeneity. The method's effectiveness is demonstrated through superior performance over existing techniques on the DFSS task, marking a significant advancement in UAV image segmentation.
{"title":"Prototypical Metric Segment Anything Model for Data-Free Few-Shot Semantic Segmentation","authors":"Zhiyu Jiang;Ye Yuan;Yuan Yuan","doi":"10.1109/LSP.2024.3476208","DOIUrl":"https://doi.org/10.1109/LSP.2024.3476208","url":null,"abstract":"Few-shot semantic segmentation (FSS) is crucial for image interpretation, yet it is constrained by requirements for extensive base data and a narrow focus on foreground-background differentiation. This work introduces Data-free Few-shot Semantic Segmentation (DFSS), a task that requires limited labeled images and forgoes the need for extensive base data, allowing for comprehensive image segmentation. The proposed method utilizes the Segment Anything Model (SAM) for its generalization capabilities. The Prototypical Metric Segment Anything Model is introduced, featuring an initial segmentation phase followed by prototype matching, effectively addressing the learning challenges posed by limited data. To enhance discrimination in multi-class segmentation, the Supervised Prototypical Contrastive Loss (SPCL) is designed to refine prototype features, ensuring intra-class cohesion and inter-class separation. To further accommodate intra-class variability, the Adaptive Prototype Update (APU) strategy dynamically refines prototypes, adapting the model to class heterogeneity. The method's effectiveness is demonstrated through superior performance over existing techniques on the DFSS task, marking a significant advancement in UAV image segmentation.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"31 ","pages":"2800-2804"},"PeriodicalIF":3.2,"publicationDate":"2024-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142438589","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：