检索结果-内蒙古大学图书馆

ExpLLM: Towards Chain of Thought for Facial Expression Recognition

ieee Transactions on Multimedia 2025年 27卷 3069-3081页

作者： Lan, Xing Xue, Jian Qi, Ji Jiang, Dongmei Lu, Ke Chua, Tat-Seng University of Chinese Academy of Sciences Beijing China Tsinghua University Beijing China Cheng Laboratory Shenzhen China National University of Singapore Singapore

Facial expression recognition (FER) is a critical task in multimedia with significant implications across various domains. However, analyzing the causes of facial expressions is essential for accurately recognizing them. Current approaches, such as those based on facial action units (AUs), typically provide AU names and intensities but lack insight into the interactions and relationships between AUs and the overall expression. In this paper, we propose a novel method called ExpLLM, which leverages large language models to generate an accurate chain of thought (CoT) for facial expression recognition. Specifically, we have designed the CoT mechanism from three key perspectives: key observations, overall emotional interpretation, and conclusion. The key observations describe the AU's name, intensity, and associated emotions. The overall emotional interpretation provides an analysis based on multiple AUs and their interactions, identifying the dominant emotions and their relationships. Finally, the conclusion presents the final expression label derived from the preceding analysis. Furthermore, we also introduce the Exp-CoT Engine, designed to construct this expression CoT and generate instruction-description data for training our ExpLLM. Extensive experiments on the RAF-DB and AffectNet datasets demonstrate that ExpLLM outperforms current state-of-the-art FER methods. ExpLLM also surpasses the latest GPT-4o in expression CoT generation, particularly in recognizing micro-expressions where GPT-4o frequently fails. © 1999-2012 ieee.

关键词： Lips Face recognition Engines Gold visualization large language models Cognition Tuning Accuracy Training

来源：评论

学校读者我要写书评

暂无评论

Extended Abstract: Quantum-Accelerated Transient Stability Assessment for Power Systems

Extended Abstract: Quantum-Accelerated Transient Stability A...

引用

ieee-Computer-Society Annual symposium on VLSI (ISVLSI)

作者： Chen, Jianing Li, Yan Penn State Univ Dept Elect Engn State Coll PA 16802 USA

ISBN: (纸本)9798350354119

Quantum machine learning (QML) methods possess the potential to address transient stability assessment (TSA) in power systems characterized by high computational complexity. In this paper, we introduce a QML-based transient stability assessment method, designed to assess the stability of modern large-scale power systems. Initially, quantum principal component analysis (QPCA) is employed to extract the most crucial features of power systems with an exponentially speedup than traditional PCA. Subsequently, to project original data into a lower-dimensional space to utilize state-of-the-art quantum computing resources, an inner product computation method is developed in a quantum manner, requiring significantly fewer measurements. The inner product results serve as inputs of a viable variational quantum algorithm (VQA) for conducting TSA. Preliminary results have shown that the QML-based TSA method can successfully determine system stability, with more details provided in the full paper. The reduced computational complexity provided by QML enables faster, potentially online analysis for TSA, thereby taking proactive action to prevent system failure.

关键词： Transient stability assessment power system dynamics quantum machine learning quantum principal component analysis variational quantum algorithm

来源：评论

学校读者我要写书评

暂无评论

DatUS: data-Driven Unsupervised Semantic Segmentation With Pretrained Self-Supervised Vision Transformer

引用

ieee TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS 2024年第5期16卷 1775-1788页

作者： Kumar, Sonal Sur, Arijit Baruah, Rashmi Dutta Indian Inst Technol Gauhati 781039 India

Successive proposals of several self-supervised training schemes (STSs) continue to emerge, taking one step closer to developing a universal foundation model. In this process, unsupervised downstream tasks are recognized as one of the evaluation methods to validate the quality of visual features learned with self-supervised training. However, unsupervised dense semantic segmentation has yet to be explored as a downstream task, which can utilize and evaluate the quality of semantic information introduced in patch-level feature representations during self-supervised training of vision transformers. Therefore, we propose a novel data-driven framework, DatUS, to perform unsupervised dense semantic segmentation (DSS) as a downstream task. DatUS generates semantically consistent pseudosegmentation masks for an unlabeled image dataset without using visual prior or synchronized data. The experiment shows that the proposed framework achieves the highest MIoU (24.90) and average F1 score (36.3) by choosing DINOv2 and the highest pixel accuracy (62.18) by choosing DINO as the STS on the training set of SUIM dataset. It also outperforms state-of-the-art methods for the unsupervised DSS task with 15.02% MIoU, 21.47% pixel accuracy, and 16.06% average F1 score on the validation set of SUIM dataset. It achieves a competitive level of accuracy for a large-scale COCO dataset.

关键词： Training Task analysis Transformers Semantic segmentation Feature extraction visualization Semantics Computer vision deep learning representation learning self-supervised learning semantic segmentation unsupervised learning vision transformer (ViT)

来源：评论

学校读者我要写书评

暂无评论

Transferring Knowledge From Text to Video: Zero-Shot Anticipation for Procedural Actions

引用

ieee TRANSACTIONS ON PATTERN analysis AND MACHINE INTELLIGENCE 2023年第6期45卷 7836-7852页

作者： Sener, Fadime Saraf, Rishabh Yao, Angela Univ Bonn D-53113 Bonn Germany Indian Inst Technol ISM Dhanbad Dhanbad 826004 Jharkhand India Natl Univ Singapore Singapore 119077 Singapore

Can we teach a robot to recognize and make predictions for activities that it has never seen before? We tackle this problem by learning models for video from text. This paper presents a hierarchical model that generalizes instructional knowledge from large-scale text corpora and transfers the knowledge to video. Given a portion of an instructional video, our model recognizes and predicts coherent and plausible actions multiple steps into the future, all in rich natural language. To demonstrate the capabilities of our model, we introduce the Tasty Videos dataset V2, a collection of 4022 recipes for zero-shot learning, recognition and anticipation. Extensive experiments with various evaluation metrics demonstrate the potential of our method for generalization, given limited video data for training models.

关键词： visualization Robots data models Task analysis Predictive models Natural languages Text recognition Deep learning action anticipation zero-shot learning video analysis

来源：评论

学校读者我要写书评

暂无评论

Lightweight Blockchain Simulation with Transaction Graph Visualizer 23

Lightweight Blockchain Simulation with Transaction Graph Vis...

引用

23rd ieee International symposium on Computational Intelligence and Informatics, CINTI 2023

作者： Sipos, Miklos Szenasi, Sandor Obuda University John von Neumann Faculty of Informatics Budapest Hungary

ISBN: (纸本)9798350342949

In this paper, a lightweight blockchain simulation and transaction graph visualization application is presented, crafted to augment the identification of pivotal nodes within blockchain networks. Harnessing sophisticated data visualization techniques, this application provides an extensive and easy to use environment for delving into and scrutinizing intricate transaction data within the blockchain. By incorporating principles from graph theory, this tool empowers users to distinguish critical nodes within the blockchain, unveiling the key participants, addresses, or entities involved in diverse transactions. This dual-functionality, featuring both blockchain simulation in C# and visualization in JavaScript, holds substantial potential for enhancing the analysis of blockchain networks. It promises to furnish valuable insights for researchers, cryptocurrency enthusiasts, and cybersecurity professionals, facilitating their endeavors to comprehend and fortify blockchain ecosystems. © 2023 ieee.

关键词： data visualization

来源：评论

学校读者我要写书评

暂无评论

Business Intelligence Tools for data Extractions Transformations and visualizations

Business Intelligence Tools for Data Extractions Transformat...

引用

2024 International symposium on Networks, Computers and Communications, ISNCC 2024

作者： Liu, Dejang College of DuPage Computer & Information Science Glen Ellyn United States

ISBN: (纸本)9798350364910

data cleansing has become an essential task not only in data analysis but also in artificial intelligence processes. As datasets dramatically increase in size, the efficiency and effectiveness of dataset preparation play a critical role in the business world. This paper reviews the features and compares the performance of data extraction, transformation, and visualization (ETV) of business intelligence tools, Tableau and Power BI. The sample datasets are 1.3 million LinkedIn job postings from the month of January 2024, as posted on Kaggle. The versions of these business intelligence tools used in this review are their newest versions at the time of preparing this report. The findings indicate that execution time to complete ETV is minutes versus hours difference these two tools even in the final visualization. © 2024 ieee.

关键词： BI Tools Power BI Tableau

来源：评论

学校读者我要写书评

暂无评论

Leveraging National Science data Fabric Services to Train data Scientists

Leveraging National Science Data Fabric Services to Train Da...

引用

2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and analysis, SC Workshops 2024

作者： Taufer, Michela Martinez, Heberth Panta, Aashish Olaya, Paula Marquez, Jack Gooch, Amy Scorzelli, Giorgio Pascucci, Valerio University of Tennessee Knoxville United States University of Utah Salt Lake City United States

ISBN: (纸本)9798350355543

We document an interactive half-day tutorial in which participants explore the advanced applications of National Science data Fabric (NSDF) services and strategies for comprehensive scientific data analysis. Targeting researchers, students, developers, and scientists, the tutorial provides valuable insights into managing and analyzing large datasets, particularly those exceeding 100TB. Participants gain hands-on experience by constructing modular workflows, leveraging public and private data storage and streaming solutions, and deploying sophisticated visualization and analysis dashboards. The tutorial emphasizes NSDF's role in supporting visualization conference themes by providing scalable visualization and visual analytics solutions. Our tutorial includes an overview of NSDF's capabilities, addressing common data analysis challenges, and intermediate hands-on exercises using NSDF services for Earth science data. Advanced applications cover handling and visualizing massive datasets requiring high-resolution data management. By the end of the session, attendees have a deeper understanding of integrating NSDF services into their research workflows, enhancing data accessibility, sharing, and collaborative scientific discovery. Our tutorial aims to advance knowledge in data-intensive computing and empower participants to harness the full potential of NSDF in their respective fields. © 2024 ieee.

关键词： data analysis data visualization workforce development

来源：评论

学校读者我要写书评

暂无评论

PyTracer: Automatically Profiling Numerical Instabilities in Python

引用

ieee TRANSACTIONS ON COMPUTERS 2023年第6期72卷 1792-1803页

作者： Chatelain, Yohan Young, Nigel Yong Sao Kiar, Gregory Glatard, Tristan Concordia Univ Dept Comp Sci & Software Engn Montreal PQ H3G 1M8 Canada Child Mind Inst Ctr Developing Brain New York NY 10022 USA

Numerical stability is a crucial requirement of reliable scientific computing. However, despite the pervasiveness of Python in data science, analyzing large Python programs remains challenging due to the lack of scalable numerical analysis tools available for this language. To fill this gap, we developed PyTracer, a profiler to quantify numerical instability in Python applications. PyTracertransparently instruments Python code to produce numerical traces and visualize them interactively in a Plotly dashboard. We designed PyTracerto be agnostic to numerical noise model, allowing for numerical profiling through Monte-Carlo Arithmetic, random rounding, random data perturbation, or structured noise for a particular application. We illustrate PyTracer's capabilities by testing the numerical stability of key functions in both SciPy and Scikit-learn, two dominant Python libraries for mathematical modeling. Through these evaluations, we demonstrate PyTraceras a scalable, automated, and generic framework for numerical profiling in Python.

关键词： Instruments Python Arithmetic Numerical stability Numerical models Libraries Codes Computer arithmetic quality analysis and evaluation stochastic analysis tracing visualization techniques and methodologies

来源：评论

学校读者我要写书评

暂无评论

Towards Robust Object Detection in Unseen Catheterization Laboratories

Towards Robust Object Detection in Unseen Catheterization La...

引用

ieee International symposium on Medical Measurements and Applications (MeMeA)

作者： Wang, Zipeng Butler, Rick van den Dobbelsteen, John J. Hendriks, Benno H. W. Van der Elst, Maarten Dauwels, Justin Delft Univ Technol Delft Netherlands Philips Res Labs Eindhoven Netherlands Reinier de Graaf Grp Delft Netherlands

ISBN: (纸本)9798350308006;9798350307993

Deep learning-based object detectors, while offering exceptional performance, are data-dependent and can suffer from generalization issues. In this work, we investigated deep neural networks for detecting people and medical instruments for the vision-based workflow analysis system inside Catheterization Laboratories (Cath Labs). The central problem explored in this paper is the fact that the performance of the detector can degrade drastically if it is trained and tested on data from different Cath Labs. Our research aimed to investigate the underlying causes of this specific performance degradation and find solutions to mitigate this issue. We employed the YOLOv8 object detector and created datasets from clinical procedures recorded at Reinier de Graaf Hospital (RdGG) and Philips Best Campus, supplemented with publicly accessible images. Through a series of experiments complemented by data visualization, we discovered that the performance degradation primarily stems from data distribution shifts in the feature space. Notably, the object detector trained on non-sensitive online images can generalize to unseen Cath Labs, outperforming the model trained on a procedure recording from a different Cath Lab. The detector trained on the online images achieved an mAP@0.5 of 0.517 on the RdGG dataset. Furthermore, by switching to the most suitable camera for each object in the Cath Lab, the multi-camera system can further improve the detection performance significantly. An aggregated 1-camera mAP@0.5 of 0.679 is achieved for single-object classes on the RdGG dataset.

关键词： Object Detection Catheterization Laboratory Domain Shift Clinical Workflow analysis

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Impact of COVID-19 on U.S. Employment Through data visualization

Analyzing the Impact of COVID-19 on U.S. Employment Through ...

引用

2024 International symposium on Networks, Computers and Communications, ISNCC 2024

作者： Ghazizadeh, Weria Ghazizadeh, Puya Old Dominion University Department of Computer Science NorfolkVA United States St. John's University Division of Computer Science Mathematics and Science New YorkNY United States

ISBN: (纸本)9798350364910

Millions of people have been impacted by the COVID-19 pandemic drastically. This pandemic has challenged different areas of people's lives, such as employment, health, the economy, education, etc. Employment rate is a significant factor in economic growth, poverty reduction, and social stability. Given the significant impact of the pandemic since early 2020, it is crucial to examine how it has affected employment rates. In this paper, we utilize data visualization to present a compre-hensive analysis that can aid government efforts in planning and managing employment during pandemics such as COVID-19. In this work, we use the collected data from the New York Times COVID-19 dataset and data from the U.S. Bureau of Labor Statistics. © 2024 ieee.

关键词： COVID-19 data visualization Employment Rate R

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：