检索结果-内蒙古大学图书馆

2025 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2025

作者： Bin Aziz, Abu Zahid Teja Karanam, Mokshagna Sai Kataria, Tushar Elhabian, Shireen Y. Scientific Computing and Imaging Institute Kahlert School of Computing University of Utah Salt Lake CityUT United States

ISBN: (纸本)9798331510831

Transformers have emerged as the state-of-the-art architecture in medical image registration, outperforming convolutional neural networks (CNNs) by addressing their limited receptive fields and overcoming gradient instability in deeper models. Despite their success, transformer-based models require substantial resources for training, including data, memory, and computational power, which may restrict their applicability for end users with limited resources. In particular, existing transformer-based 3D image registration architectures face two critical gaps that challenge their efficiency and effectiveness. Firstly, although window-based attention mechanisms reduce the quadratic complexity of full attention by focusing on local regions, they often struggle to effectively integrate both local and global information. Secondly, the granularity of tokenization, a crucial factor in registration accuracy, presents a performance tradeoff: smaller voxel-size tokens enhance detail capture but come with increased computational complexity, higher memory usage, and a greater risk of over-fitting. We present EFFICIENTMORPH, a transformer-based architecture for unsupervised 3D image registration that balances local and global attention in 3D volumes through a plane-based attention mechanism and employs a Hi-Res tokenization strategy with merging operations, thus capturing finer details without compromising computational efficiency. Notably, EFFICIENTMORPH sets a new benchmark for performance on the OASIS dataset with 16-27x fewer parameters. https://***/MedVIC-Lab/*** © 2025 IEEE.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Sparse Grid Approximation of Nonlinear SPDEs: The Landau-Lifshitz-Gilbert Equation

引用

SIAM-ASA Journal on Uncertainty Quantification 2025年第2期13卷 472-517页

作者： An, Xin Dick, Josef Feischl, Michael Scaglioni, Andrea Tran, Thanh School of Mathematics \& Statistics UNSW Sydney Australia Institute of Analysis and Scientific Computing TU Wien Vienna Austria

We show convergence rates for a sparse grid approximation of the distribution of solutions of the stochastic Landau-Lifshitz-Gilbert (LLG) equation. Beyond being a frequently studied equation in engineering and physics, the stochastic LLG equation poses many interesting challenges that do not appear simultaneously in previous works on uncertainty quantification: The equation is strongly nonlinear and time-dependent and has a nonconvex side constraint. Moreover, the parametrization of the stochastic noise features countably many unbounded parameters and low regularity compared to other elliptic and parabolic problems studied in uncertainty quantification. We use a novel technique to establish uniform holomorphic regularity of the parameter-to-solution map based on a Gronwall-type estimate and the implicit function theorem. This method is very general and is based on a set of abstract assumptions. Thus, it can be applied beyond the LLG equation as well. We demonstrate numerically the feasibility of approximating with sparse grid and show a clear advantage of a multilevel sparse grid scheme. © 2025 Society for Industrial and Applied Mathematics and American Statistical Association.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Matching Aggregate Posteriors in the Variational Autoencoder 27th

Matching Aggregate Posteriors in the Variational Autoencod...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Saha, Surojit Joshi, Sarang Whitaker, Ross Scientific Computing and Imaging Institute The University of Utah Salt Lake City United States

ISBN: (纸本)9783031781711

The variational autoencoder (VAE) [17] is a well-studied, deep, latent-variable model (DLVM) that optimizes the variational lower bound of the log marginal data likelihood. However, the VAE’s known failure to match the aggregate posterior often results unacceptable latent distribution, e.g. with pockets, holes, or clusters, that fail to adequately resemble the prior. The training of the VAE under different scenarios can also result in posterior collapse, which is associated with a loss of information in the latent space. This paper addresses these shortcomings in VAEs by reformulating the objective function to match the aggregate/marginal posterior distribution to the prior. We use kernel density estimation (KDE) to model the aggregate posterior. We propose an automated method to estimate the kernel and account for the associated kernel bias in our estimation, which enables the use of KDE in high-dimensional latent spaces. The proposed method is named the aggregate variational autoencoder (AVAE) and is built on the theoretical framework of the VAE. Empirical evaluation of the proposed method on multiple benchmark datasets demonstrates the advantages of the AVAE relative to state-of-the-art (SOTA) DLVM methods. Here is the link to the code: https://***/Surojit-Utah/AVAE. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Aggregates

来源：评论

学校读者我要写书评

暂无评论

VISTA: A Visual and Textual Attention Dataset for Interpreting Multimodal Models

VISTA: A Visual and Textual Attention Dataset for Interpreti...

引用

2025 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2025

作者： Harshit Tasdizen, Tolga School of Computing University of Utah Salt Lake CityUT United States Scientific Computing and Imaging Institute University of Utah Salt Lake CityUT United States

ISBN: (纸本)9798331536626

The recent developments in deep learning (DL) led to the integration of natural language processing (NLP) with computer vision, resulting in powerful integrated Vision and Language Models. Despite their remarkable capabilities, these models are frequently regarded as black boxes within the machine learning research community. This raises a critical question: which parts of an image correspond to specific segments of text, and how can we decipher these associations? Understanding these connections is essential for enhancing model transparency, interpretability, and trustworthiness. To answer this question, we present an image-text aligned human visual attention dataset (VISTA)11The data is available at https://***/h-pal/Data-for-VISTA that maps specific associations between image regions and corresponding text segments. We then compare the internal heatmaps generated by VL models with this dataset, allowing us to analyze and better understand the model's decision-making process. This approach aims to enhance model transparency, interpretability, and trustworthiness by providing insights into how these models align visual and linguistic information. We conducted a comprehensive study on text-guided visual saliency detection in these VL models. This study aims to understand how different models prioritize and focus on specific visual elements in response to corresponding text segments, providing deeper insights into their internal mechanisms and improving our ability to interpret their outputs. © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Beat of sound generated from NACA0012 airfoil on application of suction-blowing excitation

引用

Journal of the Acoustical Society of America 2025年第5期157卷 3730-3741页

作者： Kumar, Shashi Bhumkar, Yogesh G. Scientific Computing Laboratory School of Mechanical Sciences Indian Institute of Technology Bhubaneswar Odisha Bhubaneswar 752050 India

The present numerical study explores the effect of periodic suction-blowing excitation (SBE) on sound generation in the flow over a NACA0012 airfoil, with a focus on inducing beats in the resulting sound signals. Simulations are conducted at a Reynolds number (Re) of 5000 and an angle of attack of 5° using direct numerical simulation (DNS) approach where compressible Navier-Stokes equations are solved to compute both flow and sound fields directly. At low Re, vortex-shedding generates tonal noise at the Strouhal frequency (St). The introduction of periodic SBE adds sound sources at the excitation frequency ( f f ), with the resultant sound field arising from the interaction between vortex-shedding and excitation-induced sound sources. This study investigates the effects of excitation forcing frequency ( f f ) and amplitude ( A e ), identifying conditions that lead to beat formation. Beats are observed when f f is close to St and the sound intensities of both sources are comparable. Distinct wave propagation patterns emerge during beat formation, significantly differing from non-beat scenarios. A key contribution of this work is to offer a guideline on the relationship between excitation amplitude and the formation of sound beats, either in the axial direction of flow or normal to it. © 2025 Acoustical Society of America.

关键词： Acoustic wave propagation

来源：评论

学校读者我要写书评

暂无评论

Mesh2SSM++: A Probabilistic Framework for Unsupervised Learning of Statistical Shape Model of Anatomies from Surface Meshes

arXiv

引用

arXiv 2025年

作者： Iyer, Krithika Karanam, Mokshagna Sai Teja Elhabian, Shireen Scientific Computing and Imaging Institute The Kahlert School of Computing University of Utah UT United States

Anatomy evaluation is crucial for understanding the physiological state, diagnosing abnormalities, and guiding medical interventions. Statistical shape modeling (SSM) is vital in this process, particularly in medical image analysis and computational anatomy. By enabling the extraction of quantitative morphological shape descriptors from medical imaging data such as MRI and CT scans, SSM provides comprehensive descriptions of anatomical variations within a population. However, the effectiveness of SSM in anatomy evaluation hinges on the quality and robustness of the shape models, which face challenges due to substantial nonlinear variability in human anatomy. While deep learning techniques show promise in addressing these challenges by learning complex nonlinear representations of shapes, existing models still have limitations and often require pre-established shape models for training. To overcome these issues, we propose Mesh2SSM++, a novel approach that learns to estimate correspondences from meshes in an unsupervised manner. This method leverages unsupervised, permutation-invariant representation learning to estimate how to deform a template point cloud into subject-specific meshes, forming a correspondence-based shape model. Additionally, our probabilistic formulation allows learning a population-specific template, reducing potential biases associated with template selection. A key feature of Mesh2SSM++ is its ability to quantify aleatoric uncertainty, which captures inherent data variability and is essential for ensuring reliable model predictions and robust decision-making in clinical tasks, especially under challenging imaging conditions. Through extensive validation across diverse anatomies, evaluation metrics, and downstream tasks, we demonstrate that Mesh2SSM++ outperforms existing methods. Its ability to operate directly on meshes, combined with computational efficiency and interpretability through its probabilistic framework, makes it an attractive alter

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

Joint Audio-Visual Idling Vehicle Detection with Streamlined Input Dependencies

Joint Audio-Visual Idling Vehicle Detection with Streamlined...

引用

2025 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2025

作者： Li, Xiwen Mohammed, Rehman Mangin, Tristalee Saha, Surojit Kelly, Kerry Whitaker, Ross Tasdizen, Tolga Scientific Computing and Imaging Institute University of Utah United States University of Utah Department of Chemical Engineering United States

ISBN: (纸本)9798331536626

Idling vehicle detection (IVD) can be helpful in monitoring and reducing unnecessary idling and can be integrated into real-time systems to address the resulting pollution and harmful products. The previous approach [16], a non-end-to-end model, requires extra user clicks to specify a part of the input, making system deployment more error-prone or even not feasible. In contrast, we introduce an end-to-end joint audio-visual IVD task designed to detect vehicles visually under three states: moving, idling and engine off. Unlike feature co-occurrence task such as audio-visual vehicle tracking, our IVD task addresses complementary features, where labels cannot be determined by a single modality alone. To this end, we propose AVIVD-Net, a novel network that integrates audio and visual features through a bidirectional attention mechanism. AVIVD-Net streamlines the input process by learning a joint feature space, reducing the deployment complexity of previous methods. Additionally, we introduce the AVIVD dataset, which is seven times larger than previous datasets, offering significantly more annotated samples to study the IVD problem. Our model achieves performance comparable to prior approaches, making it suitable for automated deployment. Furthermore, by evaluating AVIVDNet on the feature co-occurrence public dataset MAVD [26], we demonstrate its potential for extension to self-driving vehicle video-camera setups. © 2025 IEEE.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

EFFICIENTMORPH: Parameter-Efficient Transformer-Based Architecture for 3D Image Registration

EFFICIENTMORPH: Parameter-Efficient Transformer-Based Archit...

引用

IEEE Workshop on Applications of Computer Vision (WACV)

作者： Abu Zahid Bin Aziz Mokshagna Sai Teja Karanam Tushar Kataria Shireen Y. Elhabian Scientific Computing and Imaging Institute & Kahlert School of Computing University of Utah Salt Lake City UT USA

ISBN: (数字)9798331510831

ISBN: (纸本)9798331510848

Transformers have emerged as the state-of-the-art architecture in medical image registration, outperforming convolutional neural networks (CNNs) by addressing their limited receptive fields and overcoming gradient instability in deeper models. Despite their success, transformer-based models require substantial resources for training, including data, memory, and computational power, which may restrict their applicability for end users with limited resources. In particular, existing transformer-based 3D image registration architectures face two critical gaps that challenge their efficiency and effectiveness. Firstly, although window-based attention mechanisms reduce the quadratic complexity of full attention by focusing on local regions, they often struggle to effectively integrate both local and global information. Secondly, the granularity of tokenization, a crucial factor in registration accuracy, presents a performance tradeoff: smaller voxel-size tokens enhance detail capture but come with increased computational complexity, higher memory usage, and a greater risk of over-fitting. We present EFFICIENTMORPH, a transformer-based architecture for unsupervised 3D image registration that balances local and global attention in 3D volumes through a plane-based attention mechanism and employs a Hi-Res tokenization strategy with merging operations, thus capturing finer details without compromising computational efficiency. Notably, EFFICIENTMORPH sets a new benchmark for performance on the OASIS dataset with ~16-27x fewer parameters. https://***/MedVIC-Lab/***

关键词： Training Image registration Three-dimensional displays Attention mechanisms Computational modeling Merging Memory management Focusing Transformers Tokenization

来源：评论

学校读者我要写书评

暂无评论

AdaSemSeg: An Adaptive Few-shot Semantic Segmentation of Seismic Facies

arXiv

引用

arXiv 2025年

作者： Saha, Surojit Whitaker, Ross Scientific Computing and Imaging Institute Kahlert School of Computing The University of Utah Salt Lake City United States

Automated interpretation of seismic images using deep learning methods is challenging because of the limited availability of training data. Few-shot learning is a suitable learning paradigm in such scenarios due to its ability to adapt to a new task with limited supervision (small training budget). Existing few-shot semantic segmentation (FSSS) methods fix the number of target classes. Therefore, they do not support joint training on multiple datasets varying in the number of classes. In the context of the interpretation of seismic facies, fixing the number of target classes inhibits the generalization capability of a model trained on one facies dataset to another, which is likely to have a different number of facies. To address this shortcoming, we propose a few-shot semantic segmentation method for interpreting seismic facies that can adapt to the varying number of facies across the dataset, dubbed the AdaSemSeg. In general, the backbone network of FSSS methods is initialized with the statistics learned from the ImageNet dataset for better performance. The lack of such a huge annotated dataset for seismic images motivates using a self-supervised algorithm on seismic datasets to initialize the backbone network. We have trained the AdaSemSeg on three public seismic facies datasets with different numbers of facies and evaluated the proposed method on multiple metrics. The performance of the AdaSemSeg on unseen datasets (not used in training) is better than the prototype-based few-shot method and baselines. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Everywhere and Nowhere: Envisioning a computing Continuum for Science

引用

computing in Science & Engineering 2025年第1期27卷 51-56页

作者： Manish Parashar Scientific Computing and Imaging Institute University of Utah Salt Lake City UT USA

Emerging data-driven scientific workflows are seeking to leverage distributed data sources to understand end-to-end phenomena, drive experimentation, and facilitate important decision making. Despite the exponential growth of available digital data sources at the edge, and the ubiquity of nontrivial computational power for processing these data, realizing such science workflows remains challenging. This article explores a computing continuum that is everywhere and nowhere: One spanning resources at the edges, in the core and in between, and providing abstractions that can be harnessed to support science. It also introduces recent research in programming abstractions that can express what data should be processed and when and where it should be processed, and autonomic middleware services that automate the discovery of resources and the orchestration of computations across these resources.

关键词： Soft sensors Decision making Programming Drives Middleware Digital information Data models Resource management

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：