检索结果-内蒙古大学图书馆

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Cao, Quanhui Tang, Pengjie Wang, Hanli Tongji Univ Dept Comp Sci & Technol Shanghai Peoples R China Jinggangshan Univ Coll Elect & Informat Engn Jian Jiangxi Peoples R China Tongji Univ Key Lab Embedded Syst & Serv Comp Minist Educ Shanghai Peoples R China Frontiers Sci Ctr Intelligent Autonomous Syst Shanghai Peoples R China

ISBN: (纸本)9781665484855

Video captioning is a sequence-to-sequence task of automatically generating descriptions for given videos. Due to the diversity of video scenes, learning rich representations is critical for video captioning. However, previous works mainly exploited elaborate features but neglected the loss of information caused by frame sampling and image compression. In this paper, we propose a novel spatio-temporal super-resolution (STSR) network which is jointly trained for the video captioning task and the video super-resolution task in an end-to-end fashion. Specifically, a video super-resolution task consists of two subtasks: spatial super-resolution restores high-resolution image features while temporal super-resolution reconstructs missing frame features between two adjacent sampled frames. By sharing multi-modal encoders across both of these two tasks, STSR encourages encoders to capture salient visual contents and learn context-aware representations. Experiments on two benchmark datasets demonstrate that the proposed STSR boosts video captioning performances significantly and outperforms most state-of-the-art approaches.

关键词： video captioning video super-resolution encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Automatic facial expressions, gaze direction and head movements generation of a virtual agent 24

Automatic facial expressions, gaze direction and head moveme...

引用

24th ACM International Conference on Multimodal Interaction (ICMI)

作者： Delbosc, Alice Ochs, Magalie Ayache, Stephane Aix Marseille Univ Marseille France

ISBN: (纸本)9781450393898

In this article, we present two models to jointly and automatically generate the head, facial and gaze movements of a virtual agent from acoustic speech features. Two architectures are explored: a Generative Adversarial Network and an Adversarial encoder-decoder. Head movements and gaze orientation are generated as 3D coordinates, while facial expressions are generated using action units based on the facial action coding system. A large corpus of almost 4 hours of videos, involving 89 different speakers is used to train our models. We extract the speech and visual features automatically from these videos using existing tools. The evaluation of these models is conducted objectively with measures such as density evaluation and a visualisation from PCA reduction, as well as subjectively through a users perceptive study. Our proposed methodology shows that on 15 seconds sequences, encoder-decoder architecture drastically improves the perception of generated behaviours in two criteria: the coordination with speech and the naturalness. Our code can be found in : https://***/aldelb/non-verbal-behaviours-generation.

关键词： Non-verbal behaviour behaviour generation embodied conversational agent neural networks adversarial learning encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

AttentionHCare: Advances on computer-aided medical prognosis using attention-based neural networks

AttentionHCare: Advances on computer-aided medical prognosis...

引用

IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) / IEEE World Congress on Computational Intelligence (IEEE WCCI) / International Joint Conference on Neural Networks (IJCNN) / IEEE Congress on Evolutionary Computation (IEEE CEC)

作者： De Barros, Pedro H. F. Rodrigues, Jose F. Univ Selo Paulo Inst Math & Comp Sci Sao Carlos SP Brazil

ISBN: (数字)9781728186719

ISBN: (纸本)9781728186719

With the increasing number of institutions adhering to Electronic Health Records (EHRs) and with the evolution of Artificial Neural Networks, the research on computer-aided medical systems has gained attention. The development of automated prognosis systems is a topic of intense research;nevertheless, despite significant advances, such systems have not yet benefited from techniques of attention and encoding-decoding. We investigate this field by designing a neural network architecture, named AttentionHCare, that innovates by putting together the latest methods for patient trajectory prediction (longitudinal prediction);that is, given a history of hospital admissions, what is the most probable clinical event to take place next? Answering that question might ease the work of healthcare professionals, who must deal with massive amounts of data when working on preventive prognosis. We proposed an architecture composed of Long Short-Term Memory (LSTM), attention mechanisms, and an encoder-decoder scheme in a single ensemble. Over datasets MIMIC-III and MIMIC-IV-ED, our experiments demonstrated results that surpassed baselines and state-of-the-art proposals. Our contribution is to elucidate the use and efficacy of the most recent techniques on neural networks in the medical domain, raising design decisions for current and future works.

关键词： Deep learning encoder-decoder Patient trajectory prediction Medical prognosis prediction Longitudinal Prediction

来源：评论

学校读者我要写书评

暂无评论

IMPROVING END-TO-END MODELS FOR SET PREDICTION IN SPOKEN LANGUAGE UNDERSTANDING 47

IMPROVING END-TO-END MODELS FOR SET PREDICTION IN SPOKEN LAN...

引用

47th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Kuo, Hong-Kwang J. Tuske, Zoltan Thomas, Samuel Kingsbury, Brian Saon, George IBM Res AI Armonk NY 10504 USA IBM Corp Armonk NY USA AppTek Mclean VA USA

ISBN: (纸本)9781665405409

The goal of spoken language understanding (SLU) systems is to determine the meaning of the input speech signal, unlike speech recognition which aims to produce verbatim transcripts. Advances in end-to-end (E2E) speech modeling have made it possible to train solely on semantic entities, which are far cheaper to collect than verbatim transcripts. We focus on this set prediction problem, where entity order is unspecified. Using two classes of E2E models, RNN transducers and attention based encoder-decoders, we show that these models work best when the training entity sequence is arranged in spoken order. To improve E2E SLU models when entity spoken order is unknown, we propose a novel data augmentation technique along with an implicit attention based alignment method to infer the spoken order. F1 scores significantly increased by more than 11% for RNN-T and about 2% for attention based encoder-decoder SLU models, outperforming previously reported results.

关键词： spoken language understanding encoder-decoder attention speech recognition ATIS

来源：评论

学校读者我要写书评

暂无评论

Two-Phase Open-Domain Question Answering System 42nd

Two-Phase Open-Domain Question Answering System

引用

42nd SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence

作者： Prasannan, Vysakh Shemshian, Shahin Gurkan, Arinc Saheer, Lakshmi Babu Oghaz, Mahdi Maktabdar Anglia Ruskin Univ Cambridge CB1 IPT England

ISBN: (纸本)9783031214400;9783031214417

Text-based Internet content is increasing at a very rapid rate day by day. As a result, even the best search engines are struggling to retrieve the exact expected results of users' queries. On many occasions, the users' expected result is embedded and scattered in a number of different documents and conventional search engines are unable to pinpoint it. To address this shortcoming, this study proposes a two-phased question answering system that utilizes a K-means clustering algorithm alongside the T5 deep encoder-decoder model to formulate a concise short answer to users' queries. The proposed system has been trained using the Kaggle QA and SQuAD datasets and achieved the maximum F1-score of 0.564 and a minimum loss of 8.56.

关键词： Natural language processing NLP K-Means Information retrieval TF-IDF encoder-decoder Question answering system

来源：评论

学校读者我要写书评

暂无评论

Machine Translation for Indian Languages Utilizing Recurrent Neural Networks and Attention

Machine Translation for Indian Languages Utilizing Recurrent...

引用

International Conference on Distributed Computing and Optimization Tech-niques (ICDCOT)

作者： Sharma, Sonali Diwakar, Manoj Graph Era Deemed Be Univ Dept CSE Dehra Dun India

ISBN: (纸本)9789811922817;9789811922800

Neural Machine Translation better known as NMT is an end-to-end approach for autonomous language translation that utilizes neural models. This is an effort to bridge the gap between the multinational and multilingual people to understand their views. The NMT systems involves models to learn directly through mapping of input-output which has proven to generate increased accuracy outputs. This technique has made remarkable accomplishments and has overcome the weakness of the conventional translations models. The paper implements the RNN, Attention based mechanism and transformer on Indian-English language pairs. So far there are no specific benchmarks for Indian languages. There are companies such as Facebook, Bing, Google whose translators supports few Indian languages. In this research work models have been trained on two set of Indian language pairs which have been retrieved from open source platform Tatoeba.

关键词： Attention mechanism encoder-decoder Machine translation Recurrent Neural Network (RNN) Transformers

来源：评论

学校读者我要写书评

暂无评论

Single Image Cloud Removal Using U-Net and Generative Adversarial Networks

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2021年第8期59卷 6371-6385页

作者： Zheng, Jiahao Liu, Xiao-Yang Wang, Xiaodong Chinese Acad Sci Shenzhen Inst Adv Technol Shenzhen 518055 Peoples R China Columbia Univ Dept Elect Engn Now York NY 10027 USA

Cloud removal is a ubiquitous and important task in remote sensing image processing, which aims at restoring the ground regions shadowed by clouds. It is challenging to remove the clouds for a single satellite image due to the difficulty of distinguishing clouds from white objects on the ground and filling the irregular missing regions with visual consistency. In this article, we propose a novel two-stage cloud removal method. The first stage is cloud segmentation, i.e., extracting the clouds and removing the thin clouds directly using U-Net. The second stage is image restoration, i.e., removing the thick cloud and recovering the corresponding irregular missing regions using generative adversarial network (GAN). We evaluate the proposed scheme on both synthetic images and real satellite images (over 20000 x 20000 pixels). On synthetic images for cloud coverage less than 40%, the proposed scheme achieves improvements of 0.049-0.078 in Structural SIMilarity (SSIM) and 3.8-6.2 dB in peak signal-to-noise ratio (PSNR), while the l(1)-norm error reduces by 49%-78%, compared with a state-of-the-art deep learning method Pix2Pix. On real satellite images, we demonstrate the consistent visual results of the proposed scheme.

关键词： encoder-decoder generative adversarial network (GAN) single image cloud removal U-Net

来源：评论

学校读者我要写书评

暂无评论

Composite load modeling by spatial-temporal deep attention network based on wide-area monitoring systems

引用

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2021年第6期40卷 12215-12226页

作者： Ghafarokhi, Omid Izadi Moattari, Mazda Forouzantabar, Ahmad Islamic Azad Univ Dept Elect & Comp Engn Marvdasht Branch Marvdasht Iran

With the development of the wide-area monitoring system (WAMS), power system operators are capable of providing an accurate and fast estimation of time-varying load parameters. This study proposes a spatial-temporal deep network-based new attention concept to capture the dynamic and static patterns of electrical load consumption through modeling complicated and non-stationary interdependencies between time sequences. The designed deep attention-based network benefits from long short-term memory (LSTM) based component to learning temporal features in time and frequency-domains as encoder-decoder based recurrent neural network. Furthermore, to inherently learn spatial features, a convolutional neural network (CNN) based attention mechanism is developed. Besides, this paper develops a loss function based on a pseudo-Huber concept to enhance the robustness of the proposed network in noisy conditions as well as improve the training performance. The simulation results on IEEE 68-bus demonstrates the effectiveness and superiority of the proposed network through comparison with several previously presented and state-of-the-art methods.

关键词： Composite load modeling deep attention neural network encoder-decoder long short-term memory convolutional neural network wide-area monitoring system

来源：评论

学校读者我要写书评

暂无评论

Lost data reconstruction for structural health monitoring using deep convolutional generative adversarial networks

引用

STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL 2021年第4期20卷 2069-2087页

作者： Lei, Xiaoming Sun, Limin Xia, Ye Tongji Univ Dept Bridge Engn Shanghai 200092 Peoples R China Tongji Univ State Key Lab Disaster Reduct Civil Engn Shanghai Peoples R China

In the application of structural health monitoring, the measured data might be temporarily or permanently lost due to sensor fault or transmission failure. The measured data with a high data loss ratio undermine its ability for modal identifications and structural condition evaluations. To reconstruct the lost data in the field of structural health monitoring, this study proposes a deep convolutional generative adversarial network which includes a generator with encoder-decoder structure and an adversarial discriminator. The proposed generative adversarial network model needs to understand the content of the complete signals, as well as produce realistic hypotheses for the lost signals. Given the data stably measured before the occurrence of data loss, the generator is trained to extract the features maintained in the data set and reconstruct lost signals using the responses of the remaining functional sensors alone. The discriminator feeds back the distinguished results to the generator to improve its reconstruction accuracy. When training the model, the reconstruction loss and the adversarial loss are employed to better handle the low-frequency features and high-frequency features of the signals. The effectiveness and efficiency of the proposed method are validated by two case studies. As the number of training epoch increases, the reconstructed signals learn the features from low-frequency to high-frequency, and the amplitude of the reconstructed signals gradually increases. It can be seen that the final reconstruction signals match well with the real signals in the time domain and frequency domain. To further demonstrate the applicability of the reconstructed signals in data analysis, the reconstructed acceleration data are used to accurately identify the modal parameters in the numerical case, and the vehicle-induced responses are precisely decomposed from the reconstructed strain data in the field case. Finally, the reconstruction capacity is also investigated

关键词： structural health monitoring data loss data reconstruction generative adversarial network encoder-decoder deep learning

来源：评论

学校读者我要写书评

暂无评论

Image Captioning with Memorized Knowledge

引用

COGNITIVE COMPUTATION 2021年第4期13卷 807-820页

作者： Chen, Hui Ding, Guiguang Lin, Zijia Guo, Yuchen Shan, Caifeng Han, Jungong Tsinghua Univ Sch Software Beijing Peoples R China Microsoft Res Beijing Peoples R China Philips Res Eindhoven Netherlands Univ Warwick WMG Data Sci Coventry W Midlands England

Image captioning, which aims to automatically generate text description of given images, has received much attention from researchers. Most existing approaches adopt a recurrent neural network (RNN) as a decoder to generate captions conditioned on the input image information. However, traditional RNNs deal with the sequence in a recurrent way, squeezing the information of all previous words into hidden cells and updating the context information by fusing the hidden states with the current word information. This may miss the rich knowledge too far in the past. In this paper, we propose a memory-enhanced captioning model for image captioning. We firstly introduce an external memory to store the past knowledge, i.e., all the information of generated words. When predicting the next word, the decoder can retrieve knowledge information about the past by means of a selective reading mechanism. Furthermore, to better explore the knowledge stored in the memory, we introduce several variants that consider different types of past knowledge. To verify the effectiveness of the proposed model, we conduct extensive experiments and comparisons on the well-known image captioning dataset MS COCO. Compared with the state-of-the-art captioning models, the proposed memory-enhanced captioning model shows a significant improvement in terms of the performance (improving 3.5% in terms of CIDEr). The proposed memory-enhanced captioning model, as demonstrated in the experiments, is more effective and superior to the state-of-the-art methods.

关键词： Image captioning Attention Memory encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：