检索结果-内蒙古大学图书馆

RoboCup Symposium

作者： Steuernagel, Lucas Maximo, Marcos R. O. A. Aeronaut Inst Technol Div Comp Sci Autonomous Computat Syst Lab LAB SCA Praca Marechal Eduardo Gomes 50Vila Acacias BR-12228900 Sao Jose Dos Campos SP Brazil

ISBN: (纸本)9783031284687;9783031284694

The RoboCup Small Size League employs cylindrical robots of 15 cm height and 18 cm diameter. Presently, most teams utilize a Kalman predictor to forecast the trajectory of other robots for better motion planning and decision making. The predictor is limited for such task, for it typically cannot generate complex movements that take into account the future actions of a robot. In this context, we introduce an encoder-decoder sequence-to-sequence neural network that outperforms the Kalman predictor in trajectory forecasting. The network consists of a Bi-LSTM encoder, an attention module and a LSTM decoder. It can predict 15 future time steps, given 30 past measurements, or 30 time steps, given 60 past observations. The proposed model is roughly 50% more performant than a Kalman predictor in terms of average displacement error and runs in less than 2 ms. We believe that our new architecture will improve our team's decision making and provide a better competitive advantage for all teams. We are looking forward to integrating it with our software pipeline and continuing our research by incorporating new training methods and new inputs to the model.

关键词： Trajectory prediction Neural networks encoder-decoder Small Size League Sequence-to-sequence

来源：评论

学校读者我要写书评

暂无评论

Learning Semantic Information from Machine Translation to Improve Speech-to-Text Translation

Learning Semantic Information from Machine Translation to Im...

引用

Asia-Pacific-Signal-and-Information-Processing-Association Annual Summit and Conference (APSIPA ASC)

作者： Deng, Pan Zhang, Jie Zhou, Xinyuan Ye, Zhongyi Zhang, Weitai Cui, Jianwei Dai, Lirong Univ Sci & Technol China USTC NERC SLIP Hefei Peoples R China iFlytek Res Hefei Peoples R China

ISBN: (纸本)9798350300673

End-to-end speech translation (ST) directly translates the source speech to the target text, following a typical encoder-decoder framework. However, it has shown that the conventional ST encoder is mainly used to extract long but locally attentive acoustic features, which may lead to a lack of global semantic features. In this work, we therefore propose to integrate a semantic decoder into the speech translation (SD-ST) model, where the semantic decoder can generate text-like features with more global semantic information analogously to the machine translation system (MT). We also investigate different strategies to ensure length consistency between text-like features and text sequences. Experimental results show that the proposed SD-ST model achieves the best BLEU score on the 40-hour subset of the Fisher Spanish English dataset and a comparable BLEU score on the MuST-C dataset. Furthermore, it is shown that the SD-ST model can even perform zero-shot ST.

关键词： End-to-end speech translation semantic information encoder-decoder speech recognition

来源：评论

学校读者我要写书评

暂无评论

A Novel Graph-Based Trajectory Predictor With Pseudo-Oracle

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2022年第12期33卷 7064-7078页

作者： Yang, Biao Yan, Guocheng Wang, Pin Chan, Ching-Yao Song, Xiang Chen, Yang Changzhou Univ Dept Informat Sci & Engn Changzhou 213000 Jiangsu Peoples R China Univ Calif Berkeley Calif PATH Richmond CA 94804 USA Nanjing Xiaozhuang Univ Sch Elect Engn Nanjing 211171 Peoples R China

Pedestrian trajectory prediction in dynamic scenes remains a challenging and critical problem in numerous applications, such as self-driving cars and socially aware robots. Challenges concentrate on capturing pedestrians' motion patterns and social interactions, as well as handling the future uncertainties. Recent studies focus on modeling pedestrians' motion patterns with recurrent neural networks, capturing social interactions with pooling- or graph-based methods, and handling future uncertainties by using the random Gaussian noise as the latent variable. However, they do not integrate specific obstacle avoidance experiences (OAEs) that may improve prediction performance. For example, pedestrians' future trajectories are always influenced by others in front. Here, we propose the Graph-based Trajectory Predictor with Pseudo-Oracle (GTPPO), an encoder-decoder-based method conditioned on pedestrians' future behaviors. Pedestrians' motion patterns are encoded with a long short-term memory unit, which introduces temporal attention to highlight specific time steps. Their interactions are captured by a graph-based attention mechanism, which draws OAE into the data-driven learning process of graph attention. Future uncertainties are handled by generating multimodal outputs with an informative latent variable. Such a variable is generated by a novel pseudo-oracle predictor, which minimizes the knowledge gap between historical and ground-truth trajectories. Finally, the GTPPO is evaluated on ETH, UCY, and Stanford Drone datasets, and the results demonstrate state-of-the-art performance. Besides, the qualitative evaluations show successful cases of handling sudden motion changes in the future. Such findings indicate that GTPPO can peek into the future.

关键词： Trajectory Uncertainty Hidden Markov models Training Encoding Dynamics Recurrent neural networks encoder-decoder graph attention network latent variable predictor social attention trajectory prediction

来源：评论

学校读者我要写书评

暂无评论

Speech Enhancement with Overlapped-Frame Information Fusion and Causal Self-Attention

Speech Enhancement with Overlapped-Frame Information Fusion ...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhang, Yuewei Zou, Huanbin Zhu, Jie Department of Electronic Engineering Shanghai Jiao Tong University Shanghai China Xiaohongshu Inc. Shanghai China

ISBN: (纸本)9798350368741

For time-frequency (TF) domain speech enhancement (SE) methods, the overlap-and-add operation in the inverse TF transformation inevitably leads to an algorithmic delay equal to the window size. However, typical causal SE systems fail to utilize the future speech information within this inherent delay, thereby limiting SE performance. In this paper, we propose an overlapped-frame information fusion scheme. At each frame index, we construct several pseudo overlapped-frames, fuse them with the original speech frame, and then send the fused results to the SE model. Additionally, we introduce a causal time-frequency-channel attention (TFCA) block to boost the representation capability of the neural network. This block parallelly processes the intermediate feature maps through self-attention-based operations in the time, frequency, and channel dimensions. Experiments demonstrate the superiority of these improvements, and the proposed SE system outperforms the current advanced methods. © 2025 IEEE.

关键词： causal self-attention encoder-decoder pseudo overlapped-frames speech enhancement

来源：评论

学校读者我要写书评

暂无评论

A SHALLOW U-NET WITH SPLIT-FUSED ATTENTION MECHANISM FOR RETINAL VESSEL SEGMENTATION 30

A SHALLOW U-NET WITH SPLIT-FUSED ATTENTION MECHANISM FOR RET...

引用

30th IEEE International Conference on Image Processing (ICIP)

作者： Bhati, Amit Jain, Samir Gour, Neha Khanna, Pritee Ojha, Aparajita Werghi, Naoufel PDPM IIITDM Dept Comp Sci & Engn Jabalpur India Khalifa Univ Dept Elect Engn & Comp Sci C2PS Abu Dhabi U Arab Emirates

ISBN: (纸本)9781728198354

Extraction of retinal vascular parts is an important task in retinal disease diagnosis. Precise segmentation of the retinal vascular pattern is challenging due to its complex structure, overlapping with other anatomical structures, and crucial thin vascular structures. In recent years, complex and heavy deep learning networks have been proposed to segment retinal blood vessels accurately. However, these methods fail to detect the thin vascular structure among different patterns of thick vessels. An attention-based novel architecture is proposed to segment the thin vasculature to address this limitation. The proposed model comprises a shallow U-Net based encoder-decoder architecture with split-fuse attention (SFA) block. The proposed SFA block enables the network to identify the placement of pixels for the tree-shaped vessel patterns at their relative position during the reconstruction phase in the decoder. The attention block aggregates low-level and high-level semantic information, improving the vessel segmentation performance. Experimentation performed on publicly available fundus datasets, DRIVE, HRF, CHASE-DB1, and STARE show that the proposed method performs better than the current state-of-the-art methods. The results demonstrate the adaptability of the proposed model for clinical applications due to its low memory footprint and better performance.

关键词： Retinal Vessel Segmentation Shallow U-Net Split Fused Attention Fully Convolution Network encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

A novel approach for detecting error measurements in a network of automatic weather stations

引用

INTERNATIONAL JOURNAL OF PARALLEL EMERGENT AND DISTRIBUTED SYSTEMS 2022年第4期37卷 425-442页

作者： Llugsi, R. El Yacoubi, S. Fontaine, A. Lupera, P. Univ Perpignan Via Domitia IMAGES ESPACE DEV Perpignan France Univ Reunion Univ Montpellier Univ Antilles ESPACE DEVUniv GuyaneIRD Montpellier France Natl Polytech Sch DETRI Quito Ecuador

In the present work, a novel methodology for error detection in automatic weather stations has been implemented. Time series acquired from two highly correlated stations with a station under analysis are utilised to obtain a 24-h air temperature forecast that allows to know if a station register erroneous measurements. Four models to obtain a reliable forecast have been analysed, auto-regressive integrated moving average, Long Short-Term Memory (LSTM), LSTM stacked and a convolutional LSTM model with uncertainty error reduction. The analysis carried out exhibits a significant success with the methodology for three stations reaching error values between 0.98 degrees C and 1.50 degrees C and correlation coefficients between 0.72 and 0.81.

关键词： LSTM time series convolutional encoder-decoder neural network walk-forward validation dropout Bayesian uncertainty

来源：评论

学校读者我要写书评

暂无评论

MP-SENet: A Speech Enhancement Model with Parallel Denoising of Magnitude and Phase Spectra 24

MP-SENet: A Speech Enhancement Model with Parallel Denoising...

引用

Interspeech Conference

作者： Lu, Ye-Xin Ai, Yang Ling, Zhen-Hua Univ Sci & Technol China Natl Engn Res Ctr Speech & Language Informat Proc Hefei Peoples R China

This paper proposes MP-SENet, a novel Speech Enhancement Network which directly denoises Magnitude and Phase spectra in parallel. The proposed MP-SENet adopts a codec architecture in which the encoder and decoder are bridged by convolution-augmented transformers. The encoder aims to encode time-frequency representations from the input noisy magnitude and phase spectra. The decoder is composed of parallel magnitude mask decoder and phase decoder, directly recovering clean magnitude spectra and clean-wrapped phase spectra by incorporating learnable sigmoid activation and parallel phase estimation architecture, respectively. Multi-level losses defined on magnitude spectra, phase spectra, short-time complex spectra, and time-domain waveforms are used to train the MP-SENet model jointly. Experimental results show that our proposed MP-SENet achieves a PESQ of 3.50 on the public VoiceBank+DEMAND dataset and outperforms existing advanced speech enhancement methods.

关键词： speech enhancement encoder-decoder parallel denoising magnitude spectra phase spectra

来源：评论

学校读者我要写书评

暂无评论

An Unsupervised Symmetric Tensor Network for Change Detection in Multitemporal Hyperspectral Images

An Unsupervised Symmetric Tensor Network for Change Detectio...

引用

Conference on Earth and Space - From Infrared to Terahertz (ESIT)

作者： Liang, Chengfang Chen, Zhao Donghua Univ Sch Comp Sci & Technol Shanghai 201620 Peoples R China

ISBN: (纸本)9781510661196;9781510661202

Since Hyperspectral images (HSIs) contain a large amount of spectral information, they can provide detailed spectral information and enable accurate CD. However, the spectral heterogeneity of HSIs may lead to false alarms which will reduce detection accuracy. Additionally, it is difficult to collect and annotate pixel-level labels for CD in HSIs. Therefore, we propose an unsupervised symmetric tensor network (USTN) for HSIs CD. We design a novel multidimensional symmetric tensor framework to solve the problem of high-dimensional data processing. Furthermore, the framework integrates a spatial edge loss to preserve detailed spectral-spatial information. Finally, we use feature fusion to suppress the invariant components (i.e., the background) and highlight the variant components (i.e., temporal changes). Experiments on two sets of multitemporal HSIs, Hermiston and Bay Area, demonstrate the effectiveness of USTN for binary change detection.

关键词： Hyperspectral images change detection unsupervised learning symmetric tensor network encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

DSCUNet: A multi-view adaptive fusion network for pavement crack segmentation 7

DSCUNet: A multi-view adaptive fusion network for pavement c...

引用

7th International Conference on Electronic Information Technology and Computer Engineering (EITCE)

作者： Luan, Shen Wang, Chen Zhang, Hongyi Gao, Xingen Mao, Yihai Xiamen Univ Technol Sch Optoelect & Commun Engn Xiamen Peoples R China

ISBN: (纸本)9798400708305

The complexity of cracks in terms of topology, shape, and background makes road crack detection based on computer vision a challenging task. In this study, we consider a network called DSCUNet, which exploits the encoder-decoder structure of U-Net for pavement crack image segmentation. A novel adaptive multi-view feature fusion module is provided in this approach. Inspired by the recently proposed dynamic snake convolutional network (DSCNet), we integrate dynamic snake convolution into the proposed model to enhance its perception of the crack structure during the feature extraction process and to produce a multi-view localized feature map. The adaptive multi-view feature fusion module is used to adaptively fuse multi-view features, aiming to capture local and global features at the same time for accurate positioning. DSCUNet is trained and tested on the CRWKH100 dataset. A detailed comparison of DSCUNet with DSCNet, U-Net, and the existing approaches is provided in our study. The results show that DSCUNet achieves finer results for crack image segmentation, achieving stateof-the-art performance with 0.843 ODS, 0.866 OIS and 0.899 AP on CRWKH100, respectively.

关键词： Crack detection encoder-decoder DSCUNet U-Net Dynamic snake convolution Segmentation

来源：评论

学校读者我要写书评

暂无评论

Multi-class Segmentation of Dark-Grown Arabidopsis Seedlings with U-Net 11

Multi-class Segmentation of Dark-Grown Arabidopsis Seedlings...

引用

11th International Conference on Bioinformatics and Computational Biology (ICBCB)

作者： Tong, Haonan Bullock, David Schmittling, Selene Haque, Samiul Tompkins, Vincent Chen, Hao Zhang, Chuang Stepanova, Anna N. Alonso, Jose M. Williams, Cranos M. Inspur Elect Informat Ind Co Ltd Beijing 100094 Peoples R China North Carolina State Univ Dept Plant & Microbial Biol Raleigh NC 27695 USA North Carolina State Univ Dept Elect & Comp Engn Raleigh NC 27695 USA

ISBN: (纸本)9798350397871

Plant organ phenotyping represents a powerful tool to investigate the effects of biotic and abiotic factors on plant growth and development. Phenotyping typically involves the initial capture of high-resolution images of the plant of interest and the subsequent measurement of different morphometric parameters of specific plant organs. This second step is typically very time consuming and difficult to automate. To deal with this bottleneck, we developed a multi-class segmentation model based on U-shape network to identify hypocotyls and roots of young seedlings of the reference plant Arabidopsis thaliana. We applied a balanced cross entropy loss function to learn an alternative optimal network structure for this multi-class segmentation task. We evaluated our segmentation machine using 66 images of the wild-type Arabidopsis strain Col-0, as well as 34 images of the highly agravitropic and morphologically distinct mutant strain, aux1-7. Our model achieved a mean BFscore of 0.81 for Col-0 seedlings and 0.75 for aux1-7 mutant seedlings on the test dataset. Our model was also able to maintain accuracy in these two morphologically different genotypes suggesting that our segmentation procedure could be successfully applied to Arabidopsis seedlings showing broad morphological differences due to their genotype or treatment conditions. Appropriate segmentation is the first step in identifying phenotypic changes under hormone-mediated stress response. The identified growth parameters will be useful to identify the response associated with both abiotic and biotic stresses, which include but not limited to drought stress, heat stress, and the presence of pests or pathogens. The quantification of these parameters will aid assessment of genetic factors that contribute to the stress response.

关键词： Multi-class segmentation plant phenotype U-Net encoder-decoder deep learning

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：