Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674072
Chenyi Qiang, Li Ping, Amin Ul Haq, Liu He, Abdul Haq
Net traffic classification is the basis for providing various network services such as network security, network monitoring and Quality of Service (QoS) etc. Therefore, this field has always been a hot spot in academic and industrial research. Through proper data processing, the researcher said that it is possible to classify network flows through machine learning. Therefore, it is necessary to explore suitable processing methods and model structures. To our best knowledge, classification methods based on sequential feature learning are rarely discussed, so this paper proposes a model based on sequential features of net traffic. Different from the previous classification methods based on machine learning, the classification method based on GRU network focuses on exploring the sequential feature information of network traffic. This classification model based on deep learning is very suitable for big data processing. In terms of evaluation, we used the USTC-TFC2016 data set, compared with the basic model and previous methods, the experimental results show that: (1) the effectiveness of the sequential model for net traffic classification. (2) The sequential model has good performance in both accuracy and stability.
{"title":"Net Traffic Classification Based on GRU Network Using Sequential Features","authors":"Chenyi Qiang, Li Ping, Amin Ul Haq, Liu He, Abdul Haq","doi":"10.1109/ICCWAMTIP53232.2021.9674072","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674072","url":null,"abstract":"Net traffic classification is the basis for providing various network services such as network security, network monitoring and Quality of Service (QoS) etc. Therefore, this field has always been a hot spot in academic and industrial research. Through proper data processing, the researcher said that it is possible to classify network flows through machine learning. Therefore, it is necessary to explore suitable processing methods and model structures. To our best knowledge, classification methods based on sequential feature learning are rarely discussed, so this paper proposes a model based on sequential features of net traffic. Different from the previous classification methods based on machine learning, the classification method based on GRU network focuses on exploring the sequential feature information of network traffic. This classification model based on deep learning is very suitable for big data processing. In terms of evaluation, we used the USTC-TFC2016 data set, compared with the basic model and previous methods, the experimental results show that: (1) the effectiveness of the sequential model for net traffic classification. (2) The sequential model has good performance in both accuracy and stability.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124020283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674151
D. Kulevome, Wang Hong, Xue-gang Wang, Bernard M. Cobbinah, B. L. Y. Agbley, Qurratulain Safder
The ability of a system to operate within its acceptable range under various operating conditions is essential in determining its reliability. In this paper, a health management framework is developed for the continuous assessment and confidence in the operational fidelity of a given radar system. The complexity of such a system makes it challenging to implement a comprehensive prognostics and health management approach effectively. For this reason, the system is analyzed at the subsystem level to consider components' degeneration within each subsystem. In the proposed framework, sensors are used to collect relevant data from critical components for system diagnosis. Subsequent preprocessing and analysis can then be used in developing a degradation model and efficient decision-making process.
{"title":"System Diagnosis Framework for Sustaining the Operational Fidelity of a Radar System","authors":"D. Kulevome, Wang Hong, Xue-gang Wang, Bernard M. Cobbinah, B. L. Y. Agbley, Qurratulain Safder","doi":"10.1109/ICCWAMTIP53232.2021.9674151","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674151","url":null,"abstract":"The ability of a system to operate within its acceptable range under various operating conditions is essential in determining its reliability. In this paper, a health management framework is developed for the continuous assessment and confidence in the operational fidelity of a given radar system. The complexity of such a system makes it challenging to implement a comprehensive prognostics and health management approach effectively. For this reason, the system is analyzed at the subsystem level to consider components' degeneration within each subsystem. In the proposed framework, sensors are used to collect relevant data from critical components for system diagnosis. Subsequent preprocessing and analysis can then be used in developing a degradation model and efficient decision-making process.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"2005 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128856417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674065
Isaac Adjei-Mensah, Xiaoling Zhang, Adu Asare Baffour, Isaac Osei Agyemang, S. B. Yussif, B. L. Y. Agbley, Collins Sey
Vision Transformers use self-attention techniques to learn long-range spatial relations to focus on the relevant parts of an image. They have achieved state-of-the-art results in many computer vision tasks. Recently, some methods have to leverage Vision Transformer-based models to tackle tasks in medical imaging. However, Vision Transformer emphasizes the low-resolution features due to the repetitive downsamplings, which result in a loss or lack of detailed localization information, making it highly unfit for low-level image recognition. In this paper, we investigate the performance of Vision Transformer on low-level medical images and contrast it with convolutional neural networks. The experimental results show that Convolutional Neural Network outperforms the Vision Transformer-based models on all four datasets.
{"title":"Investigating Vision Transformer Models for Low-Resolution Medical Image Recognition","authors":"Isaac Adjei-Mensah, Xiaoling Zhang, Adu Asare Baffour, Isaac Osei Agyemang, S. B. Yussif, B. L. Y. Agbley, Collins Sey","doi":"10.1109/ICCWAMTIP53232.2021.9674065","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674065","url":null,"abstract":"Vision Transformers use self-attention techniques to learn long-range spatial relations to focus on the relevant parts of an image. They have achieved state-of-the-art results in many computer vision tasks. Recently, some methods have to leverage Vision Transformer-based models to tackle tasks in medical imaging. However, Vision Transformer emphasizes the low-resolution features due to the repetitive downsamplings, which result in a loss or lack of detailed localization information, making it highly unfit for low-level image recognition. In this paper, we investigate the performance of Vision Transformer on low-level medical images and contrast it with convolutional neural networks. The experimental results show that Convolutional Neural Network outperforms the Vision Transformer-based models on all four datasets.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128566451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674118
Abena Achiaa Atwereboannah, Wei-Ping Wu, Lei Ding, S. B. Yussif, Edwin Kwadwo Tenagyei
Protein-ligand prediction plays a key role in drug discovery. Nevertheless, many algorithms are over reliant on 3D structure representations of proteins and ligands which are often rare. Techniques that can leverage the sequence-level representations of proteins, ligands and pockets are thus required to predict binding affinity and facilitate the drug discovery process. We have proposed a deep learning model with an attention mechanism to predict protein-ligand binding affinity. Our model is able to make comparable achievements with state-of-the-art deep learning models used for protein-ligand binding affinity prediction.
{"title":"Protein-Ligand Binding Affinity Prediction Using Deep Learning","authors":"Abena Achiaa Atwereboannah, Wei-Ping Wu, Lei Ding, S. B. Yussif, Edwin Kwadwo Tenagyei","doi":"10.1109/ICCWAMTIP53232.2021.9674118","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674118","url":null,"abstract":"Protein-ligand prediction plays a key role in drug discovery. Nevertheless, many algorithms are over reliant on 3D structure representations of proteins and ligands which are often rare. Techniques that can leverage the sequence-level representations of proteins, ligands and pockets are thus required to predict binding affinity and facilitate the drug discovery process. We have proposed a deep learning model with an attention mechanism to predict protein-ligand binding affinity. Our model is able to make comparable achievements with state-of-the-art deep learning models used for protein-ligand binding affinity prediction.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125946683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674121
Chen Sijie, Qing Peijun, H. Mei, Xiong Yuting, Su Caijun, Liu Zhongyong, Qin Peiyun
Tens of millions of tons of desulfurized gypsum can be produced in China every year. Desulfurization gypsum is used as an admixture to replace part of cement in concrete. And the smaller the particle size of desulfurized gypsum, the better the mechanical properties of cement mortar. The compressive and flexural tests show that the cement mortar with 3% and 4% desulfurized gypsum has the best mechanical properties, and the mechanical properties of superfine desulfurized gypsum are better than those of direct desulfurized gypsum.
{"title":"Superfine Desulfurization Gypsum can Improve the Strength of Cement Mortar","authors":"Chen Sijie, Qing Peijun, H. Mei, Xiong Yuting, Su Caijun, Liu Zhongyong, Qin Peiyun","doi":"10.1109/ICCWAMTIP53232.2021.9674121","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674121","url":null,"abstract":"Tens of millions of tons of desulfurized gypsum can be produced in China every year. Desulfurization gypsum is used as an admixture to replace part of cement in concrete. And the smaller the particle size of desulfurized gypsum, the better the mechanical properties of cement mortar. The compressive and flexural tests show that the cement mortar with 3% and 4% desulfurized gypsum has the best mechanical properties, and the mechanical properties of superfine desulfurized gypsum are better than those of direct desulfurized gypsum.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"IA-19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126560699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674097
Abdulrahman Al-Malahi, Omar Almaqtari, W. Ayedh, B. Tang
Due to the density of the crowded electromagnetic environment nowadays and the complexity of modern radar signals, the performance of pulse repetition interval (PRI)-based sorting systems experience more deterioration than ever before. Such systems are considered unreliable when working in crowded circumstances, moreover, they require a long pulse stream and high signal-to-noise (SNR) ratio, which makes obtaining acceptable sorting accuracy a difficult task. In this paper, a new machine learning architecture, Combined Residual and Recurrent Neural Network (CRRNN), is proposed, where recurrent neural network (RNN) and residual neural network (ResNet) are incorporated to create an architecture which can be used to overcome the above-mentioned shortcomings of conventional sorting methods achieving more accuracy and stability. Separate ResNet and RNN models are investigated as well for comparison. Simulations are performed after discussion of the structure and the principle of work of each network architecture. Statistical results showing the high and reliable performance of the proposed method in different conditions are presented and discussed.
{"title":"Radar Signal Sorting Using Combined Residual and Recurrent Neural Network (CRRNN)","authors":"Abdulrahman Al-Malahi, Omar Almaqtari, W. Ayedh, B. Tang","doi":"10.1109/ICCWAMTIP53232.2021.9674097","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674097","url":null,"abstract":"Due to the density of the crowded electromagnetic environment nowadays and the complexity of modern radar signals, the performance of pulse repetition interval (PRI)-based sorting systems experience more deterioration than ever before. Such systems are considered unreliable when working in crowded circumstances, moreover, they require a long pulse stream and high signal-to-noise (SNR) ratio, which makes obtaining acceptable sorting accuracy a difficult task. In this paper, a new machine learning architecture, Combined Residual and Recurrent Neural Network (CRRNN), is proposed, where recurrent neural network (RNN) and residual neural network (ResNet) are incorporated to create an architecture which can be used to overcome the above-mentioned shortcomings of conventional sorting methods achieving more accuracy and stability. Separate ResNet and RNN models are investigated as well for comparison. Simulations are performed after discussion of the structure and the principle of work of each network architecture. Statistical results showing the high and reliable performance of the proposed method in different conditions are presented and discussed.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125025229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674120
Gan Tao, Li Weichao, He Yanmin, Luo Yu
Few-shot learning (FSL) is a technique for learning new class concepts with a limited number of labeled samples, is a key step towards human-level intelligence. Among existing few-shot learning methods, prototypical network shows to be promising in solving the critical problem of overfitting. However, due to the simplicity of average operation in building the prototype representation for each class, the inter- and intra-class relationships among the samples in the support set are not fully exploited, resulting in deviation of the prototype representation from the true class distribution. In this paper, we propose graph-based prototypical network (GPN) model to overcome this problem. In GPN, a fully learnable message passing graph module is proposed to refine the feature embedding vector of each sample. The refined features are then fed into prototypical network to obtain the robust prototype representations of classes. According to experimental results, the proposed method achieves competitive classification accuracy against state-of-the-art ones.
{"title":"Graph-Based Prototypical Network for Few-Shot Learning","authors":"Gan Tao, Li Weichao, He Yanmin, Luo Yu","doi":"10.1109/ICCWAMTIP53232.2021.9674120","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674120","url":null,"abstract":"Few-shot learning (FSL) is a technique for learning new class concepts with a limited number of labeled samples, is a key step towards human-level intelligence. Among existing few-shot learning methods, prototypical network shows to be promising in solving the critical problem of overfitting. However, due to the simplicity of average operation in building the prototype representation for each class, the inter- and intra-class relationships among the samples in the support set are not fully exploited, resulting in deviation of the prototype representation from the true class distribution. In this paper, we propose graph-based prototypical network (GPN) model to overcome this problem. In GPN, a fully learnable message passing graph module is proposed to refine the feature embedding vector of each sample. The refined features are then fed into prototypical network to obtain the robust prototype representations of classes. According to experimental results, the proposed method achieves competitive classification accuracy against state-of-the-art ones.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122165913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the last decade, the application of pedestrian detection in computer vision has gradually increased, such as social distance detection in the epidemic era. In this paper, we improve the newly proposed YOLOv5 model, use the idea of deep mutual learning for training, compare the performance and accuracy of different parameters, and select a relatively good model. As for the application, after detecting an abnormal pedestrian or a designated pedestrian, we use the Deep SORT method to track the pedestrian via the pedestrian's ID. Experimental analysis shows that our model performs well in terms of mean average precision (mAP), total loss (TL), and frames per second (FPS).
{"title":"Pedestrian Detection and Tracking with Deep Mutual Learning","authors":"Feng Xudong, Guo Xiaofeng, Kuang Ping, Liao Xianglong, Zhu Yalou","doi":"10.1109/ICCWAMTIP53232.2021.9674099","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674099","url":null,"abstract":"In the last decade, the application of pedestrian detection in computer vision has gradually increased, such as social distance detection in the epidemic era. In this paper, we improve the newly proposed YOLOv5 model, use the idea of deep mutual learning for training, compare the performance and accuracy of different parameters, and select a relatively good model. As for the application, after detecting an abnormal pedestrian or a designated pedestrian, we use the Deep SORT method to track the pedestrian via the pedestrian's ID. Experimental analysis shows that our model performs well in terms of mean average precision (mAP), total loss (TL), and frames per second (FPS).","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131395201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674098
Huang Ying, Deng Fumin, Chen Yujie
This research analyses the impact of product placement on brand trust in mobile communication product-harm crisis and establishes a conceptual model of how product placement affects brand trust based on perceived risk and relationship distance as mediators. Based on the Bass Diffusion Model, simulation is adopted to analyze the changing trend of the number of people corresponding to each variable after the crisis occurs and after product placement is added. The study found weak correlation between product placement and brand trust and strong correlation between the mediators and brand trust, which proves that there lies a suppression effect in the impact of product placement on brand trust in the situation.
{"title":"Simulation Research on the Impact of Product Placement on Brand Trust in Mobile Communication Product-Harm Crisis","authors":"Huang Ying, Deng Fumin, Chen Yujie","doi":"10.1109/ICCWAMTIP53232.2021.9674098","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674098","url":null,"abstract":"This research analyses the impact of product placement on brand trust in mobile communication product-harm crisis and establishes a conceptual model of how product placement affects brand trust based on perceived risk and relationship distance as mediators. Based on the Bass Diffusion Model, simulation is adopted to analyze the changing trend of the number of people corresponding to each variable after the crisis occurs and after product placement is added. The study found weak correlation between product placement and brand trust and strong correlation between the mediators and brand trust, which proves that there lies a suppression effect in the impact of product placement on brand trust in the situation.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122312476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-17DOI: 10.1109/ICCWAMTIP53232.2021.9674087
Huang Yonghao, Yan Qi, Wang Xiaofang
With the advent of the Internet era, the cost of realizing virtual characters has been greatly reduced. The high cost and low efficiency exhibited by traditional virtual technology can't meet social needs. People are seeking fast, convenient and accurate virtual character reproduction technology. Through researching the characteristics of the characters appearance, language habits, voice tone and so on, the research direction of this paper is to simulate and reshape voice and images, construct a cloud platform-based on user-side and client-side, and integrate deep learning, natural language processing, digital twins and other technologies to an immersive 4D intelligent interactive platform. The platform under in the form of application software provides integrated services of intelligent voice interaction and virtual character interaction. In the industrial diagnosis mode, the transition from traditional video retention and voice retention to a new intelligent voice recognition and simulation mode is realized.
{"title":"Immersive 4D Intelligent Interactive Platform Based on Deep Learning","authors":"Huang Yonghao, Yan Qi, Wang Xiaofang","doi":"10.1109/ICCWAMTIP53232.2021.9674087","DOIUrl":"https://doi.org/10.1109/ICCWAMTIP53232.2021.9674087","url":null,"abstract":"With the advent of the Internet era, the cost of realizing virtual characters has been greatly reduced. The high cost and low efficiency exhibited by traditional virtual technology can't meet social needs. People are seeking fast, convenient and accurate virtual character reproduction technology. Through researching the characteristics of the characters appearance, language habits, voice tone and so on, the research direction of this paper is to simulate and reshape voice and images, construct a cloud platform-based on user-side and client-side, and integrate deep learning, natural language processing, digital twins and other technologies to an immersive 4D intelligent interactive platform. The platform under in the form of application software provides integrated services of intelligent voice interaction and virtual character interaction. In the industrial diagnosis mode, the transition from traditional video retention and voice retention to a new intelligent voice recognition and simulation mode is realized.","PeriodicalId":358772,"journal":{"name":"2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123122816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：