Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.04.005
Haoyang Qi , Xin Song , Yuqi Zhang , Lanfeng Li , Zhiao Cao
With the popularization of sensor technology, researchers extract Channel State Information (CSI) from WiFi, which reflects the movement of family members by observing changes in signal transmission in residences. However, WiFi signals are still affected by multipath effects in residences. Therefore, we construct a multi-scale CSI combination mask matrix between adjacent scales based on the ternary closure. Next, we propose a competitive localization network based on aggregated affinity propagation algorithm. Experiments have been conducted to demonstrate that the proposed algorithm achieves significant improvement compared to other classical algorithms in the indoor environment.
{"title":"Muloc: Multi-scale combination mask indoor localization network for WiFi based on channel state information","authors":"Haoyang Qi , Xin Song , Yuqi Zhang , Lanfeng Li , Zhiao Cao","doi":"10.1016/j.icte.2025.04.005","DOIUrl":"10.1016/j.icte.2025.04.005","url":null,"abstract":"<div><div>With the popularization of sensor technology, researchers extract Channel State Information (CSI) from WiFi, which reflects the movement of family members by observing changes in signal transmission in residences. However, WiFi signals are still affected by multipath effects in residences. Therefore, we construct a multi-scale CSI combination mask matrix between adjacent scales based on the ternary closure. Next, we propose a competitive localization network based on aggregated affinity propagation algorithm. Experiments have been conducted to demonstrate that the proposed algorithm achieves significant improvement compared to other classical algorithms in the indoor environment.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 529-535"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144296814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.01.010
Yi-Hao Tu, Yi-Wei Ma
This study introduces the Multi-Agent, Multi-Parameter, Interaction-Driven Contention Window Optimization (M2I-CWO) algorithm, a novel Multi-Agent Deep Reinforcement Learning (MADRL) framework designed to optimize multiple CW parameters in IEEE 802.11 Wireless LANs. Unlike single-parameter or specialized multi-agent methods, M2I-CWO employs a Dueling-DQN architecture and an Adaptive Interaction Reward Function—spanning independent, cooperative, competitive, and mixed modes—and accommodates Hierarchical Multi-Agent System (HMAS) or Federated RL (FRL) for further scalability. First, multiple CW parameters are simultaneously adjusted to enhance collision management. Second, M2I-CWO consistently achieves throughput improvements in both static and dynamic scenarios. Extensive results confirm M2I-CWO's superiority in efficiency and adaptability.
{"title":"A comprehensive multi-agent deep reinforcement learning framework with adaptive interaction strategies for contention window optimization in IEEE 802.11 Wireless LANs","authors":"Yi-Hao Tu, Yi-Wei Ma","doi":"10.1016/j.icte.2025.01.010","DOIUrl":"10.1016/j.icte.2025.01.010","url":null,"abstract":"<div><div>This study introduces the Multi-Agent, Multi-Parameter, Interaction-Driven Contention Window Optimization (M2I-CWO) algorithm, a novel Multi-Agent Deep Reinforcement Learning (MADRL) framework designed to optimize multiple CW parameters in IEEE 802.11 Wireless LANs. Unlike single-parameter or specialized multi-agent methods, M2I-CWO employs a Dueling-DQN architecture and an Adaptive Interaction Reward Function—spanning independent, cooperative, competitive, and mixed modes—and accommodates Hierarchical Multi-Agent System (HMAS) or Federated RL (FRL) for further scalability. First, multiple CW parameters are simultaneously adjusted to enhance collision management. Second, M2I-CWO consistently achieves throughput improvements in both static and dynamic scenarios. Extensive results confirm M2I-CWO's superiority in efficiency and adaptability.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 473-480"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.04.011
Heecheol Yang
In wireless communication systems, accurate channel state information plays a fundamental role in achieving optimal transmission efficiency at the base station (BS). We introduce a deep learning-based channel prediction designed to address the challenges posed by imperfect channel reciprocity in time-division duplex multiple-input multiple-output systems. We propose two models that not only facilitate accurate channel prediction but also perform channel calibration that can alleviate the impact of imperfect channel reciprocity between BS and users. We evaluate the performance through the simulations in line-of-sight and non-line-of-sight scenarios, demonstrating efficacy in enhancing the accuracy of predicted future downlink channels.
{"title":"Deep learning-based channel prediction for TDD MIMO systems with imperfect channel reciprocity","authors":"Heecheol Yang","doi":"10.1016/j.icte.2025.04.011","DOIUrl":"10.1016/j.icte.2025.04.011","url":null,"abstract":"<div><div>In wireless communication systems, accurate channel state information plays a fundamental role in achieving optimal transmission efficiency at the base station (BS). We introduce a deep learning-based channel prediction designed to address the challenges posed by imperfect channel reciprocity in time-division duplex multiple-input multiple-output systems. We propose two models that not only facilitate accurate channel prediction but also perform channel calibration that can alleviate the impact of imperfect channel reciprocity between BS and users. We evaluate the performance through the simulations in line-of-sight and non-line-of-sight scenarios, demonstrating efficacy in enhancing the accuracy of predicted future downlink channels.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 590-596"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.03.003
Naor Zohar
Traditionally, User-cell Association (UA) schemes for cellular networks are based solely on the quality and strength of the received signal. This mechanism may not always be adequate for the complex nature of fifth generation and beyond (B5G) networks, and may lead to load biasing. Additional parameters, such as the load on the neighboring cells, the needs of the user equipment (UE), and the UE mobility should be considered as well. The realization that relying solely on the signal strength and quality for UA may violate load balancing has been recognized for a long time. However, the approach taken was to balance the load by a network-dependent mechanism. Yet, the UA mechanism remains based on the best-received signal. The underlying assumption that all the users have the same needs remains. Load-aware UA was considered only upon mobility-driven handover to small cells. This study suggests a UE-based threshold scheme for load-aware UA that is suitable for B5G networks. Since the optimal UA problem is known to be NP-hard, we suggest a heuristic mechanism, that is sufficiently simple and reliable to be implemented in practice, yet sufficiently efficient to significantly outperform the existing UA mechanism. Simulation results demonstrated that the suggested UA scheme can potentially reduce the required network bandwidth resources by up to 25% and significantly reduce the service blocking probability and the average waiting time for service.
{"title":"To associate or not to associate? A user-based threshold scheme for 5G and beyond networks","authors":"Naor Zohar","doi":"10.1016/j.icte.2025.03.003","DOIUrl":"10.1016/j.icte.2025.03.003","url":null,"abstract":"<div><div>Traditionally, User-cell Association (UA) schemes for cellular networks are based solely on the quality and strength of the received signal. This mechanism may not always be adequate for the complex nature of fifth generation and beyond (B5G) networks, and may lead to load biasing. Additional parameters, such as the load on the neighboring cells, the needs of the user equipment (UE), and the UE mobility should be considered as well. The realization that relying solely on the signal strength and quality for UA may violate load balancing has been recognized for a long time. However, the approach taken was to balance the load by a network-dependent mechanism. Yet, the UA mechanism remains based on the best-received signal. The underlying assumption that all the users have the same needs remains. Load-aware UA was considered only upon mobility-driven handover to small cells. This study suggests a UE-based threshold scheme for load-aware UA that is suitable for B5G networks. Since the optimal UA problem is known to be NP-hard, we suggest a heuristic mechanism, that is sufficiently simple and reliable to be implemented in practice, yet sufficiently efficient to significantly outperform the existing UA mechanism. Simulation results demonstrated that the suggested UA scheme can potentially reduce the required network bandwidth resources by up to 25%<!--> <!-->and significantly reduce the service blocking probability and the average waiting time for service.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 390-401"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This research introduces an innovative approach in implementing Know Your Customer (KYC) on blockchain technology as a means of using data, hybrid robust m-estimation, and the iteratively reweighted less squares (IRLS) method to optimize CLV data. This approach aims to improve the accuracy and reliability of CLV predictions by ensuring the security and reliability of customer data. This tool can help companies manage and increase CLV more effectively, meeting data security and compliance standards. The R-squared validation test results are close to 1, so the model can explain data variations well. RMSE and MSE have small values, so the model has good performance in predicting the target value. With these achievements, this approach contributes to the development of better marketing strategies and business decisions in an increasingly complex and rapidly changing digital environment.
{"title":"Blockchain-enabled KYC integration for CLV optimization with robust M-Estimation and IRLS method","authors":"Marischa Elveny , Mahyuddin K.M. Nasution , Fanindia Purnamasari , Tengku Siti Meriam Tengku Wook","doi":"10.1016/j.icte.2025.03.006","DOIUrl":"10.1016/j.icte.2025.03.006","url":null,"abstract":"<div><div>This research introduces an innovative approach in implementing Know Your Customer (KYC) on blockchain technology as a means of using data, hybrid robust m-estimation, and the iteratively reweighted less squares (IRLS) method to optimize CLV data. This approach aims to improve the accuracy and reliability of CLV predictions by ensuring the security and reliability of customer data. This tool can help companies manage and increase CLV more effectively, meeting data security and compliance standards. The R-squared validation test results are close to 1, so the model can explain data variations well. RMSE and MSE have small values, so the model has good performance in predicting the target value. With these achievements, this approach contributes to the development of better marketing strategies and business decisions in an increasingly complex and rapidly changing digital environment.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 402-410"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.02.012
Brij B. Gupta , Akshat Gaurav , Wadee Alhalabi , Varsha Arya , Eman Alharbi , Kwok Tai Chui
The rapid growth of IoT devices has heightened the risk of botnet attacks, calling for scalable and distributed detection solutions. In this context, this study proposes a distributed optimization system for IoT attack detection using CNN model utilizing federated learning. After optimizing the hyperparameters of the model at the server, the Siberian Tiger Optimization (STO) method distributes these values to clients for dispersed training. Our model achieves accuracy, recall, and precision of 0.89978, 0.94355, and 0.94455, respectively, using the N-BaIoT dataset. These findings show, in spite of latency issues, the efficiency of federated learning in distributed IoT security systems.
{"title":"Distributed optimization for IoT attack detection using federated learning and Siberian Tiger optimizer","authors":"Brij B. Gupta , Akshat Gaurav , Wadee Alhalabi , Varsha Arya , Eman Alharbi , Kwok Tai Chui","doi":"10.1016/j.icte.2025.02.012","DOIUrl":"10.1016/j.icte.2025.02.012","url":null,"abstract":"<div><div>The rapid growth of IoT devices has heightened the risk of botnet attacks, calling for scalable and distributed detection solutions. In this context, this study proposes a distributed optimization system for IoT attack detection using CNN model utilizing federated learning. After optimizing the hyperparameters of the model at the server, the Siberian Tiger Optimization (STO) method distributes these values to clients for dispersed training. Our model achieves accuracy, recall, and precision of 0.89978, 0.94355, and 0.94455, respectively, using the N-BaIoT dataset. These findings show, in spite of latency issues, the efficiency of federated learning in distributed IoT security systems.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 542-546"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144296816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.03.011
Masoumeh Mohammadi, Insoo Sohn
This study addresses the challenge of state of health (SOH) estimation for lithium-ion batteries using a generative graphical approach under adversarial conditions. We analyze the impact of adversarial data poisoning attacks on SOH prediction models, specifically employing the fast gradient sign method (FGSM) and iterative fast gradient sign method (IFGSM). To enhance model robustness, we propose a two-defense strategy against such attacks. The effectiveness of these defenses is evaluated using error metrics such as root-mean-square error (RMSE), mean absolute error (MAE), and mean-square error (MSE). Results indicate that the proposed strategy significantly improves the model’s ability to accurately predict SOH, even in the presence of malicious data.
{"title":"Adversarial defense for battery state-of-health prediction models","authors":"Masoumeh Mohammadi, Insoo Sohn","doi":"10.1016/j.icte.2025.03.011","DOIUrl":"10.1016/j.icte.2025.03.011","url":null,"abstract":"<div><div>This study addresses the challenge of state of health (SOH) estimation for lithium-ion batteries using a generative graphical approach under adversarial conditions. We analyze the impact of adversarial data poisoning attacks on SOH prediction models, specifically employing the fast gradient sign method (FGSM) and iterative fast gradient sign method (IFGSM). To enhance model robustness, we propose a two-defense strategy against such attacks. The effectiveness of these defenses is evaluated using error metrics such as root-mean-square error (RMSE), mean absolute error (MAE), and mean-square error (MSE). Results indicate that the proposed strategy significantly improves the model’s ability to accurately predict SOH, even in the presence of malicious data.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 436-441"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297523","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.03.007
Harashta Tatimma Larasati , Byung-Soo Choi
As the works on quantum computing have seen substantial development in recent years, we are on the verge of seeing the fruitful result of the preliminary, early version of quantum computers. Nevertheless, in order to achieve full-fledged, large-scale quantum computers, two aspects still lacking in the existing quantum computers, i.e., scalability and reliability, will need to be more carefully considered and investigated. In this survey, we present a review of existing literature that aims to alleviate the scalability and reliability issues in quantum computers. In particular, we discuss how existing research leads to two main solutions: for advancing scalability, a distributed quantum computing (DQC) paradigm will be the primary direction instead of the general centralized quantum computing (CQC), whereas for attaining reliability, a fault-tolerant quantum computation (FTQC) approach would need to be leveraged instead of the existing noisy intermediate-scale quantum computers (NISQ). Combining both solutions, we highlight the essentiality of fault-tolerant distributed quantum computation (FT-DQC) and present related progress in the field. Furthermore, from the papers that we have gathered, we provide the taxonomy of the works to give a clearer landscape of the field and discuss the key issues in realizing FT-DQC.
{"title":"Towards fault-tolerant distributed quantum computation (FT-DQC): Taxonomy, recent progress, and challenges","authors":"Harashta Tatimma Larasati , Byung-Soo Choi","doi":"10.1016/j.icte.2025.03.007","DOIUrl":"10.1016/j.icte.2025.03.007","url":null,"abstract":"<div><div>As the works on quantum computing have seen substantial development in recent years, we are on the verge of seeing the fruitful result of the preliminary, early version of quantum computers. Nevertheless, in order to achieve full-fledged, large-scale quantum computers, two aspects still lacking in the existing quantum computers, i.e., scalability and reliability, will need to be more carefully considered and investigated. In this survey, we present a review of existing literature that aims to alleviate the scalability and reliability issues in quantum computers. In particular, we discuss how existing research leads to two main solutions: for advancing scalability, a distributed quantum computing (DQC) paradigm will be the primary direction instead of the general centralized quantum computing (CQC), whereas for attaining reliability, a fault-tolerant quantum computation (FTQC) approach would need to be leveraged instead of the existing noisy intermediate-scale quantum computers (NISQ). Combining both solutions, we highlight the essentiality of fault-tolerant distributed quantum computation (FT-DQC) and present related progress in the field. Furthermore, from the papers that we have gathered, we provide the taxonomy of the works to give a clearer landscape of the field and discuss the key issues in realizing FT-DQC.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 417-435"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper investigates the secure transmission problem using a dynamically reconfigurable intelligent surface (RIS). Considering an eavesdropping scenario where a multi-antenna eavesdropper has inadequate channel state information, a resilient beamforming strategy is proposed to simultaneously optimize the active beamforming at the base station and rotational angles of the RIS in order to improve the system's secrecy rate. A continuous-time propagation model is used to explain the dynamically rotating RIS to maximize the cascaded channel gain. Simulation findings validate the security benefits of the resilient beamforming approach, and the dynamic rotation technique increases the secrecy rate while reducing delay spread.
{"title":"Robust and secure communications using dynamic rotations reconfigurable intelligent surface","authors":"Kunyu Li, Guoping Zhang, Hongbo Xu, Ruijie Li, Yun Chen, Xingxing Huang","doi":"10.1016/j.icte.2025.03.010","DOIUrl":"10.1016/j.icte.2025.03.010","url":null,"abstract":"<div><div>This paper investigates the secure transmission problem using a dynamically reconfigurable intelligent surface (RIS). Considering an eavesdropping scenario where a multi-antenna eavesdropper has inadequate channel state information, a resilient beamforming strategy is proposed to simultaneously optimize the active beamforming at the base station and rotational angles of the RIS in order to improve the system's secrecy rate. A continuous-time propagation model is used to explain the dynamically rotating RIS to maximize the cascaded channel gain. Simulation findings validate the security benefits of the resilient beamforming approach, and the dynamic rotation technique increases the secrecy rate while reducing delay spread.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 523-528"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-06-01DOI: 10.1016/j.icte.2025.04.004
Hyeonjin Kim, Hyukjoon Lee
Abstract
We present a novel 3D face reconstruction framework called Facial action unit (AU) feature-based 3D FAce Reconstruction using Transformer (AUFART) that can generate a 3D face model that is responsive to AU activation given a single monocular 2D image to capture expressions. We propose a novel 3D face reconstruction framework, called AUFART (Facial Action Unit Feature-based 3D Face Reconstruction using Transformer), which generates 3D face models responsive to AU activations from a single monocular 2D image, effectively capturing facial expressions. AUFART leverages AU-specific features as well as facial global features to achieve accurate 3D reconstruction of facial expressions using transformers. We also introduce a loss function designed to guide the learning process so that the discrepancy in AU activations between the input and rendered reconstruction is minimized. The proposed framework achieves an average F1 score of 0.39, outperforming state-of-the-art methods.
{"title":"A new method for 3D face reconstruction using transformers based on action unit features","authors":"Hyeonjin Kim, Hyukjoon Lee","doi":"10.1016/j.icte.2025.04.004","DOIUrl":"10.1016/j.icte.2025.04.004","url":null,"abstract":"<div><div>Abstract</div><div>We present a novel 3D face reconstruction framework called Facial action unit (AU) feature-based 3D FAce Reconstruction using Transformer (AUFART) that can generate a 3D face model that is responsive to AU activation given a single monocular 2D image to capture expressions. We propose a novel 3D face reconstruction framework, called AUFART (Facial Action Unit Feature-based 3D Face Reconstruction using Transformer), which generates 3D face models responsive to AU activations from a single monocular 2D image, effectively capturing facial expressions. AUFART leverages AU-specific features as well as facial global features to achieve accurate 3D reconstruction of facial expressions using transformers. We also introduce a loss function designed to guide the learning process so that the discrepancy in AU activations between the input and rendered reconstruction is minimized. The proposed framework achieves an average F1 score of 0.39, outperforming state-of-the-art methods.</div></div>","PeriodicalId":48526,"journal":{"name":"ICT Express","volume":"11 3","pages":"Pages 454-459"},"PeriodicalIF":4.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297526","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号