检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Wang, Tao Yuan, Li Chen, Yunpeng Feng, Jiashi Yan, Shuicheng Institute of Data Science National University of Singapore Singapore Yitu Technology Integrative Science and Engineering Programme NUS Graduate School National University of Singapore Singapore Department of Electrical and Computer Engineering National University of Singapore Singapore

Recently, DETR [3] pioneered the solution of vision tasks with transformers, it directly translates the image feature map into the object detection result. Though effective, translating the full feature map can be costly due to redundant computation on some area like the background. In this work, we encapsulate the idea of reducing spatial redundancy into a novel poll and pool (PnP) sampling module, with which we build an end-to-end PnP-DETR architecture that adaptively allocates its computation spatially to be more efficient. Concretely, the PnP module abstracts the image feature map into fine foreground object feature vectors and a small number of coarse background contextual feature vectors. The transformer models information interaction within the fine-coarse feature space and translates the features into the detection result. Moreover, the PnP-augmented model can instantly achieve various desired trade-offs between performance and computation with a single model by varying the sampled feature length, without requiring to train multiple models as existing methods. Thus it offers greater flexibility for deployment in diverse scenarios with varying computation constraint. We further validate the generalizability of the PnP module on panoptic segmentation and the recent transformer-based image recognition model ViT [7] and show consistent efficiency gain. We believe our method makes a step for efficient visual analysis with transformers, wherein spatial redundancy is commonly observed. Code will be available at https://***/twangnh/pnp-detr. Copyright © 2021, The Authors. All rights reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Time-Domain Heart Sound Classification Using Neural Audio Codecs

Time-Domain Heart Sound Classification Using Neural Audio Co...

引用

International Conference of Advanced Informatics: Concept, Theory and Application (ICAICTA)

作者： Ansh Mishra Jia Qi Yip Eng Siong Chng Department of Physics IIT Delhi New Delhi India Interdisciplinary Graduate Programme Nanyang Technological University (NTU) Singapore School of Computer Science and Engineering Nanyang Technological University (NTU) Singapore

ISBN: (数字)9798331520311

ISBN: (纸本)9798331520328

Heart auscultation remains a valuable approach for early detection of valvular heart disease, but its efficacy depends on physician skill. The rise of telehealth services necessitates consideration of heart sound transmission and classification through compressed formats. While neural audio codecs offer impressive compression rates, they may introduce distortions that could impact classification accuracy. We propose a novel approach to perform heart sound classification directly on the continuous representations of the codec, rather than on reconstructed audio. Our method demonstrates good performance on the heart sound classification task while leveraging the encoder of the audio codec, potentially improving robustness and efficiency in telehealth applications. Using the Descript Audio Codec encoder combined with a modified M5 convolutional neural network, our approach achieves a Classification Error Rate of 0.78 on the Yaseen Dataset, outperforming models trained on transmitted audio and comparing favorably with state-of-the-art methods. This performance, coupled with the method's inherent robustness to transmission artifacts, highlights its potential for improving remote cardiac diagnostics in telehealth settings.

关键词： Heart Codecs Error analysis Telemedicine Computational modeling Medical services Robustness Real-time systems Time-domain analysis Informatics

来源：评论

学校读者我要写书评

暂无评论

Robust and Transferable Backdoor Attacks Against Deep Image Compression With Selective Frequency Prior

arXiv

引用

arXiv 2024年

作者： Yu, Yi Wang, Yufei Yang, Wenhan Guo, Lanqing Lu, Shijian Duan, Ling-Yu Tan, Yap-Peng Kot, Alex C. Lab Interdisciplinary Graduate Programme Nanyang Technological University Singapore Pengcheng Laboratory Shenzhen China School of Electrical and Electronic Engineering Nanyang Technological University Singapore School of Computer Science and Engineering Nanyang Technological University Singapore School of Computer Science Peking University Beijing China

Recent advancements in deep learning-based compression techniques have demonstrated remarkable performance surpassing traditional methods. Nevertheless, deep neural networks have been observed to be vulnerable to backdoor attacks, where an added pre-defined trigger pattern can induce the malicious behavior of the models. In this paper, we propose a novel approach to launch a backdoor attack with multiple triggers against learned image compression models. Drawing inspiration from the widely used discrete cosine transform (DCT) in existing compression codecs and standards, we propose a frequency-based trigger injection model that adds triggers in the DCT domain. In particular, we design several attack objectives that are adapted for a series of diverse scenarios, including: 1) attacking compression quality in terms of bit-rate and reconstruction quality;2) attacking task-driven measures, such as face recognition and semantic segmentation in downstream applications. To facilitate more efficient training, we develop a dynamic loss function that dynamically balances the impact of different loss terms with fewer hyper-parameters, which also results in more effective optimization of the attack objectives with improved performance. Furthermore, we consider several advanced scenarios. We evaluate the resistance of the proposed backdoor attack to the defensive pre-processing methods and then propose a two-stage training schedule along with the design of robust frequency selection to further improve resistance. To strengthen both the cross-model and cross-domain transferability on attacking downstream CV tasks, we propose to shift the classification boundary in the attack loss during training. Extensive experiments also demonstrate that by employing our trained trigger injection models and making slight modifications to the encoder parameters of the compression model, our proposed attack can successfully inject multiple backdoors accompanied by their corresponding triggers int

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Case Study of Service Oriented Enterprise Architecture Framework

引用

computer Technology and Application 2012年第3期3卷 240-249页

作者： Nassir Dino Awel Dico Dida Midekso Worku Alemu Software Engineering Track IT Doctoral Program Addis A baba University Addis A baba P. O. BOX 1176 Ethiopia Graduate Programme HiLCoE School of Computer Science ＆ Technology Addis Ababa P.O.BOX 33465 Ethiopia

Many organizations have now adopted Service Oriented Architecture （SOA） as an architectural style to help them with architecture, design and implementation of their core services and systems. Most of these organizations are challenged in integrating SOA style with their overall Enterprise Architecture work. This framework links an SOA style with the Enterprise Architecture （EA） methodologies to help organizations organize their SOA effort as a key part of their Enterprise Architecture. The case study demonstrates the implementation of architecture goal with organization vision in service oriented organizational structure using services that align Business with Technology. The framework is validated and has reserved the privileges of SOA and EA.

关键词： SOA (service oriented architecture) EA (enterprise architecture) service oriented enterprise (SOE) SOE strategicarchitecture SOE domain architecture SOE solution architecture SOE lifecycle service oriented enterprise architecture (SOEA)framework.

来源：评论

学校读者我要写书评

暂无评论

Hybrid Global Flexible Power Point Tracking Algorithm for Photovoltaic Systems

Hybrid Global Flexible Power Point Tracking Algorithm for Ph...

引用

2024 Energy Conversion Congress and Expo Europe, ECCE Europe 2024

作者： Kumaresan, Anusha Tafti, Hossein Dehghani Farivar, Glen G. Pou, Josep Strunz, Kai Niyato, Dusit Nanyang Technological University Energy Research Institute at Ntu Interdisciplinary Graduate Programme Singapore Australia The University of Melbourne Department of Electrical and Electronic Engineering VIC3010 Australia Nanyang Technological University School of Electrical and Electronic Engineering Singapore Technische Universität Chair for Sustainable Electric Networks and Sources of Energy Berlin Germany Nanyang Technological University School of Computer Science and Engineering Singapore

ISBN: (纸本)9798350364446

Frequency support in grid-connected photovoltaic (PV) systems is achieved by employing flexible power point tracking (FPPT) algorithms. Subsequently, according to the grid frequency requirements, the PV power is curtailed to a power reference, unlike the traditional method of operating the PV at the maximum power point. In practical cases, the PV arrays are prone to partial shading, and hence, there is a requirement to formulate the FPPT algorithms that operates effectively during such cases. Moreover, fast transient response is a necessity, which implies that the algorithms proposed in this paper is desirable. This paper proposes a hybrid global FPPT (GFPPT) algorithm based on secant and Newton's method, to shift the PV operating point to the required power reference. The Newton's method offers quadratic convergence rate and the secant method guarantees convergence. Hence, a hybrid algorithm consisting of these two methods will provide both fast and guaranteed convergence. Validation through simulation is presented in this paper and the proposed method is compared with the recently presented perturb and observe (P&O) GFPPT algorithm in the literature. © 2024 IEEE.

关键词： Newton-Raphson method

来源：评论

学校读者我要写书评

暂无评论

Audio-Visual Deception Detection: DOLOS Dataset and Parameter-Efficient Crossmodal Learning

Audio-Visual Deception Detection: DOLOS Dataset and Paramete...

引用

International Conference on computer Vision (ICCV)

作者： Xiaobao Guo Nithish Muthuchamy Selvaraj Zitong Yu Adams Wai-Kin Kong Bingquan Shen Alex Kot Rapid-Rich Object Search (ROSE) Lab Interdisciplinary Graduate Programme Nanyang Technological University School of Computer Science and Engineering NTU School of Electrical & Electronic Engineering NTU DSO National Laboratories Singapore

Deception detection in conversations is a challenging yet important task, having pivotal applications in many fields such as credibility assessment in business, multimedia anti-frauds, and custom security. Despite this, deception detection research is hindered by the lack of high-quality deception datasets, as well as the difficulties of learning multimodal features effectively. To address this issue, we introduce DOLOS 1 , the largest gameshow deception detection dataset with rich deceptive conversations. DOLOS includes 1, 675 video clips featuring 213 subjects, and it has been labeled with audio-visual feature annotations. We provide train-test, duration, and gender protocols to investigate the impact of different factors. We benchmark our dataset on previously proposed deception detection approaches. To further improve the performance by fine-tuning fewer parameters, we propose Parameter-Efficient Crossmodal Learning (PECL), where a Uniform Temporal Adapter (UT-Adapter) explores temporal attention in transformer-based architectures, and a crossmodal fusion module, Plug-in Audio-Visual Fusion (PAVF), combines crossmodal information from audio-visual features. Based on the rich fine-grained audio-visual annotations on DOLOS, we also exploit multi-task learning to enhance performance by concurrently predicting deception and audiovisual features. Experimental results demonstrate the desired quality of the DOLOS dataset and the effectiveness of the PECL. The DOLOS dataset and the source codes are available at here.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Structured Multi-Track Accompaniment Arrangement via Style Prior Modelling

arXiv

引用

arXiv 2023年

作者： Zhao, Jingwei Xia, Gus Wang, Ziyu Wang, Ye Institute of Data Science NUS Singapore School of Computing NUS Singapore Integrative Sciences and Engineering Programme NUS Graduate School Singapore Machine Learning Department MBZUAI United Arab Emirates Computer Science Department NYU Shanghai China

In the realm of music AI, arranging rich and structured multi-track accompaniments from a simple lead sheet presents significant challenges. Such challenges include maintaining track cohesion, ensuring long-term coherence, and optimizing computational efficiency. In this paper, we introduce a novel system that leverages prior modelling over disentangled style factors to address these challenges. Our method presents a two-stage process: initially, a piano arrangement is derived from the lead sheet by retrieving piano texture styles;subsequently, a multi-track orchestration is generated by infusing orchestral function styles into the piano arrangement. Our key design is the use of vector quantization and a unique multi-stream Transformer to model the long-term flow of the orchestration style, which enables flexible, controllable, and structured music generation. Experiments show that by factorizing the arrangement task into interpretable sub-stages, our approach enhances generative capacity while improving efficiency. Additionally, our system supports a variety of music genres and provides style control at different composition hierarchies. We further show that our system achieves superior coherence, structure, and overall arrangement quality compared to existing baselines. © 2023, CC BY.

关键词： Vector quantization

来源：评论

学校读者我要写书评

暂无评论

MoRe-Fi: Motion-robust and fine-grained respiration monitoring via deep-learning UWB radar

arXiv

引用

arXiv 2021年

作者： Zheng, Tianyue Chen, Zhe Zhang, Shujie Cai, Chao Luo, Jun School of Computer Science and Engineering Nanyang Technological University Singapore Energy Research Institute Interdisciplinary Graduate Programme Nanyang Technological University Singapore China-Singapore International Joint Research Institute Guangzhou China

Crucial for healthcare and biomedical applications, respiration monitoring often employs wearable sensors in practice, causing inconvenience due to their direct contact with human bodies. Therefore, researchers have been constantly searching for contact-free alternatives. Nonetheless, existing contact-free designs mostly require human subjects to remain static, largely confining their adoptions in everyday environments where body movements are inevitable. Fortunately, radio-frequency (RF) enabled contact-free sensing, though suffering motion interference inseparable by conventional filtering, may offer a potential to distill respiratory waveform with the help of deep learning. To realize this potential, we introduce MoRe-Fi to conduct fine-grained respiration monitoring under body movements. MoRe-Fi leverages an IR-UWB radar to achieve contact-free sensing, and it fully exploits the complex radar signal for data augmentation. The core of MoRe-Fi is a novel variational encoder-decoder network;it aims to single out the respiratory waveforms that are modulated by body movements in a non-linear manner. Our experiments with 12 subjects and 66-hour data demonstrate that MoRe-Fi accurately recovers respiratory waveform despite the interference caused by body movements. We also discuss potential applications of MoRe-Fi for pulmonary disease diagnoses. Copyright © 2021, The Authors. All rights reserved.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

In-class use of the nu-case Mobile Telehealth System in a Medical School

In-class use of the nu-case Mobile Telehealth System in a Me...

引用

IEEE International Symposium on computer-Based Medical Systems

作者： Marcelo Ballaben Carloni Manuel Cesario Raquel Rangel Cesario Saturnino Luz Edson Margarido Masood Masoodian Daniel Facciolo Pires Medical School on Health Promotion University of Franca S?o Paulo Brazil Graduate Programme on Health Promotion University of Franca S?o Paulo Brazil School of Computer Science & Statistics Trinity College Dublin Dublin Ireland Department of Computer Science The University of Waikato Hamilton New Zealand

ISBN: (纸本)9781479910526

As telehealth systems are increasingly being adopted by healthcare providers, a growing number of medical practitioners need to learn to use such systems. Despite this, however, most medical schools do not currently include telehealth-related content in their curriculum. In this paper we demonstrate, through examples from our own experience, how the Problem Based Learning method can be used to incorporate telehealth systems in medical science education. We also present a survey of medical students and faculty, conducted at a university in Brazil, to gain the participants' opinions on the nu-case mobile telehealth system. The results of the survey show that nu-case provides support for a range of tasks that a telehealth system needs to cater for in the context of our application area.

关键词： Mobile Telehealth System

来源：评论

学校读者我要写书评

暂无评论

A Formal Introduction to Batch-Integrated Gradients for Temporal Explanations

A Formal Introduction to Batch-Integrated Gradients for Temp...

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Jamie Duell Monika Seisenberger Tianlong Zhong Hsuan Fu Xiuyi Fan School of Mathematics and Computer Science Swansea University Energy Research Institute Interdisciplinary Graduate Programme Nanyang Technological University Department of Finance Insurance and Real Estate Université Laval Lee Kong Chian School of Medicine Nanyang Technological University School of Computer Science and Engineering Nanyang Technological University

eXplainable Artificial Intelligence (XAI) is at the forefront of Artificial Intelligence research. Little attention, however, has been paid to the development of XAI methods for temporal data. Current state-of-the-art methods treat instances as independent and do not utilise the time dimension. Critical fields such as Healthcare and Finance often take a temporal form, leaving a prominent gap in XAI research. To this end, we propose the utilisation and optimisation of path based methods to use the temporal nature of data for explanations. In this work we (1) Extrapolate on a new technique for explainability forming a formal introduction, based on Integrated Gradients, a technique not designed for temporal data, (2) introduce new properties for time-based explainers and give an overview of the state-of-the-art methods and their adherence to these properties, (3) provide a theoretical and empirical analysis of path based methods, (4) demonstrate explanations on real world case studies. From this, we identify the proposed method best adheres to both the proposed properties and existing properties. Similarly, we demonstrate how the introduced method outperforms state-of the-art methods in the demonstrated areas

关键词：

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：