检索结果-内蒙古大学图书馆

2024 SIGGRAPH Asia conference

作者： Zhang, Cheng Wang, Yuanhao Carrasco, Francisco Vicente Wu, Chenglei Yang, Jinlong Beeler, Thabo De la Torre, Fernando Carnegie Mellon Univ Pittsburgh PA 15213 USA Google Inc Mountain View CA USA Google Inc Zurich Switzerland

ISBN: (纸本)9798400711312

We introduce FabricDiffusion, a method for transferring fabric textures from a single clothing image to 3D garments of arbitrary shapes. Existing approaches typically synthesize textures on the garment surface through 2D-to-3D texture mapping or depth-aware inpainting via generative models. Unfortunately, these methods often struggle to capture and preserve texture details, particularly due to challenging occlusions, distortions, or poses in the input image. Inspired by the observation that in the fashion industry, most garments are constructed by stitching sewing patterns with flat, repeatable textures, we cast the task of clothing texture transfer as extracting distortion-free, tileable texture materials that are subsequently mapped onto the UV space of the garment. Building upon this insight, we train a denoising diffusion model with a large-scale synthetic dataset to rectify distortions in the input texture image. This process yields a flat texture map that enables a tight coupling with existing Physically-Based Rendering (PBR) material generation pipelines, allowing for realistic relighting of the garment under various lighting conditions. We show that FabricDiffusion can transfer various features from a single clothing image including texture patterns, material properties, and detailed prints and logos. Extensive experiments demonstrate that our model significantly outperforms state-to-the-art methods on both synthetic data and real-world, in-the-wild clothing images while generalizing to unseen textures and garment shapes.

关键词： Texture transfer BRDF material diffusion model synthetic data 3D garments reconstruction

来源：评论

学校读者我要写书评

暂无评论

Development of a New image reconstruction Method Using Bayesian Estimation with Limited View-Angle Projection data for BNCT-SPECT

引用

APPLIED SCIENCES-BASEL 2024年第20期14卷 9411页

作者： Lu, Fan Inamoto, Haruka Takeishi, Shuto Tamaki, Shingo Kusaka, Sachie Murata, Isao Osaka Univ Grad Sch Engn Div Sustainable Energy & Environm Engn Suita 5650871 Japan

Featured Application This research presents a novel image reconstruction method for Boron Neutron Capture Therapy (BNCT)-Single Photon Emission Computed Tomography (SPECT) using Bayesian estimation with limited view-angle projection data. The method aims to address the inherent challenges in BNCT-SPECT imaging where conventional algorithms struggle with incomplete data from restricted projection angles. By improving the image accuracy under such conditions, this technique has potential applications in clinical settings, where accurate tumor localization and dose distribution are critical for the success of BNCT treatments. This method could lead to enhanced treatment planning and monitoring, ultimately improving patient *** Boron Neutron Capture Therapy (BNCT) is an emerging radiation treatment for cancer, and its challenges are being explored. Systems capable of capturing real-time observations of this treatment's effectiveness, particularly BNCT-SPECT methods that measure gamma rays emitted instantaneously from outside the body during nuclear reactions and that reconstruct images using Single Photon Emission Computed Tomography (SPECT) techniques, remain unavailable. BNCT-SPECT development is hindered by two main factors, the first being the projection angle. Unlike conventional SPECT, the projection angle range which is achievable by rotating a detector array cannot exceed approximately 90 degrees. Consequently, Fourier-based image reconstruction methods, requiring projections from at least 180 degrees, do not apply to BNCT-SPECT. The second limitation is the measurement time. Given these challenges, we developed a new sequential approximation image reconstruction method using Bayesian estimation, which is effective under the stringent BNCT-SPECT conditions. We also compared the proposed method with the existing Maximum Likelihood-Expectation Maximization (ML-EM) image reconstruction method. Numerical experiments were conducted by obtaining BNCT-SPE

关键词： BNCT-SPECT image reconstruction limited view-angle projection Bayesian estimation

来源：评论

学校读者我要写书评

暂无评论

Streamlined Models of CMOS image Sensors Carbon Impacts 27

Streamlined Models of CMOS Image Sensors Carbon Impacts

引用

27th Euromicro conference on Digital System Design

作者： Weppe, Olivier Chossat, Jerome Marty, Thibaut Prevotet, Jean-Christophe Pelcat, Maxime Univ Rennes INSA Rennes CNRS IETRUMR 6164 F-35000 Rennes France STMicroelectronics AMS Imaging Div Grenoble France

ISBN: (纸本)9798350380392;9798350380385

With the escalating concern about global warming, the environmental impact of electronic devices must be scrutinized. Life Cycle Assessments (LCA) reveal that Integrated Circuits (ICs) are the primary contributors to greenhouse gas emissions in these devices. However, performing an inventory to determine the ICs impact is a complex task due to missing data and the existing studies on ICs have been neglecting CMOS image Sensors (CIS). Despite the surge in CIS usage, particularly in smartphones, there is a lack of comprehensive models to assess their environmental impact. This paper proposes a multi-level set of models that leverage available information while considering the specificities of CIS. The most comprehensive model incorporates factors such as the total silicon area, geographical location (influencing the energy mix), and the technology node. To accommodate scenarios with incomplete data, subsequent models are designed to effectively utilize averaged parameters. The proposed models are applied to sensors manufactured by STMicroelectronics and Sony, and the results are compared with existing LCA results from Fairphone. Our approach provides a more comprehensive understanding of the environmental impact of CIS, contributing to the broader goal of reducing the carbon footprint of electronic devices. Our results suggest that the carbon impact of a Fairphone 4 image sensor is likely higher than previously estimated, with a significant gap between our findings and the expected value.

关键词： Carbon Footprint CMOS image Sensors (CIS) Environmental Impact Greenhouse Gas Emissions Integrated Circuits (ICs) Life Cycle Assessment (LCA)

来源：评论

学校读者我要写书评

暂无评论

A comparative study of data-driven modal decomposition analysis of unforced and forced cylinder wakes

引用

JOURNAL OF VISUALIZATION 2023年第4期26卷 755-777页

作者： Chang, Xu Gao, Donglai Harbin Inst Technol Minist Ind & Informat Technol Key Lab Smart Prevent & Mitigat Civil Engn Disaste Harbin 150090 Peoples R China Harbin Inst Technol Minist Educ Key Lab Struct Dynam Behav & Control Harbin 150090 Peoples R China

The present study on the recognition of coherent structures in flow fields was conducted using three typical data-driven modal decomposition methods: proper orthogonal decomposition (POD), dynamic mode decomposition (DMD), and Fourier mode decomposition (FMD). Two real circular cylinder wake flows (forced and unforced), obtained from two-dimensional particle image velocimetry (2D PIV) measurements, were analyzed to extract the coherent structures. It was found that the POD method could be used to extract the large-scale structures from the fluctuating velocity in a wake flow, the DMD method showed potential for dynamical mode frequency identification and linear reconstruction of the flow field, and the FMD method provided a significant computational efficiency advantage when the dominant frequency of the flow field was known. The limitations of the three methods were also identified: The POD method was incomplete in the spatial-temporal decomposition and each mode mixed multiple frequencies leading to unclear physics, the DMD method is based on the linear assumption and thus the highly nonlinear part of the flow field was unsuitable, and the FMD method is based on global power spectrum analysis while being overwhelmed by an unknown high-frequency flow field.

关键词： Proper orthogonal decomposition (POD) Dynamic mode decomposition (DMD) Fourier mode decomposition (FMD) Vortex dynamics

来源：评论

学校读者我要写书评

暂无评论

Robust Multilayer Matrix Factorization for Blind Hyperspectral image Unmixing 16

Robust Multilayer Matrix Factorization for Blind Hyperspectr...

引用

16th International conference on Signal Processing Systems, ICSPS 2024

作者： Yang, Gang Southwest China Institute of Electronic Technology Chengdu China

ISBN: (纸本)9781510689251

Hyperspectral image unmixing estimates a collection of constituent materials (called endmembers) and their corresponding proportions (called abundances), which is a critical preprocessing step in many remote sensing applications. In this paper, we propose a novel robust multilayer matrix factorization (RMLMF) approach for hyperspectral image unmixing. Firstly, instead of sampling all observed pixels, the endmembers are modeled as a linear combination of candidate endmembers to reduce the redundant spectra and alleviate heavy computational cost. Secondly, to explore the hidden information, the shallow decomposition is extended into multilayer structure. Furthermore, considering that hyperspectral images (HSIs) still suffer from a variety of noises, L2,1-norm is employed to measure the reconstruction error of the multilayer structure. Experimental results on both synthetic and real data sets demonstrate the superiority of the proposed RMLMF over many state-of-the-art algorithms. © 2025 SPIE.

关键词： Matrix factorization

来源：评论

学校读者我要写书评

暂无评论

Dual Feature reconstruction Network For Few-shot image Classification

Dual Feature Reconstruction Network For Few-shot Image Class...

引用

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

作者： Guo, Xiaowei Wu, Jijie Ren, Kai Song, Qi Li, Xiaoxu Lanzhou Univ Technol Lanzhou Peoples R China

ISBN: (纸本)9798350300673

Few-shot image classification aims to provide accurate predictions for novelty by learning from a limited number of samples. Classical few-shot image classification methods usually use data augmentation and self-supervision to compensate for the lack of training sample, and introduce migration learning and meta-learning to pre-train the model or accelerate the model optimization, which improves the classification performance of the model. However, with a small amount of labeled sample data, these methods cannot meet the requirements of the model's ability to characterize sample features, resulting in a model that is highly susceptible to overfitting problems. In this paper, we propose a Dual Feature reconstruction Network (DFRN) for few-shot image classification. The network constructs the double feature vector by two modules, in which the first-level feature module generates an attention mask based on the image to make the feature vector characterize more of the target region, and the secondary feature module interferes with the feature vector to improve its generalization performance. In addition, the network also enhances the classification performance of the model by considering the contextual information of the support classes through an auxiliary loss function. Through extensive experiments, the network proposed in this paper achieves excellent performance on Flowers, CUB and Cars datasets and outperforms other reference fine-grained image classification methods such as FRN.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

Leveraging Classic Deconvolution and Feature Extraction in Zero-Shot image Restoration

Leveraging Classic Deconvolution and Feature Extraction in Z...

引用

IEEE/CVF International conference on Computer Vision (ICCV)

作者： Chobola, Tomas Mueller, Gesine Dausmann, Veit Theileis, Anton Taucher, Jan Huisken, Jan Peng, Tingying Tech Univ Munich Munich Germany Georg August Univ Gottingen Gottingen Germany GEOMAR Helmholtz Ctr Ocean Res Kiel Kiel Germany German Res Ctr Environm Hlth Helmholtz Munich Helmholtz AI Neuherberg Germany

ISBN: (纸本)9798350307443

Non-blind deconvolution aims to restore a sharp image from its blurred counterpart given an obtained kernel. Existing deep neural architectures are often built based on large datasets of sharp ground truth images and trained with supervision. Sharp, high quality ground truth images, however, are not always available, especially for biomedical applications. This severely hampers the applicability of current approaches in practice. In this paper, we propose a novel non-blind deconvolution method that leverages the power of deep learning and classic iterative deconvolution algorithms. Our approach combines a pre-trained network to extract deep features from the input image with iterative Richardson-Lucy deconvolution steps. Subsequently, a zero-shot optimisation process is employed to integrate the deconvolved features, resulting in a high-quality reconstructed image. By performing the preliminary reconstruction with the classic iterative deconvolution method, we can effectively utilise a smaller network to produce the final image, thus accelerating the reconstruction whilst reducing the demand for valuable computational resources. Our method demonstrates significant improvements in various real-world applications non-blind deconvolution tasks.

关键词： deconvolution microscopy self supervised zero shot

来源：评论

学校读者我要写书评

暂无评论

A deep-learning based algorithm for image reconstruction in Compton tomography 12

A deep-learning based algorithm for image reconstruction in ...

引用

12th International conference on image Processing Theory, Tools and Applications, IPTA 2023

作者： Ayad, Ishak Tarpau, Cecilia Cebeiro, Javier Nguyen, Mai K. Cy Cergy Paris Université Etis Umr 8051 and Agm Umr 8088 Cergy-Pontoise France Heriot-Watt University Edinburgh School of Mathematical and Computer Sciences United Kingdom Comision de Investigaciones Cientificas CIC Iteca UNSAM-CONICET Provincia de Buenos Aires Argentina Cy Cergy Paris Université Etis Umr 8051 Cergy-Pontoise France

ISBN: (纸本)9798350325416

In this paper we study the applications of deep-learning to the problem of image reconstruction in Compton scatter tomography, a field where deep-learning techniques are still unexplored. Particularly, we focus on a new design with uncollimated detectors that simplifies some previous configurations of Compton scanners. The system inherits attractive advantages such as non-moving components and the ability to combine with other imaging modes. Since there is no an analytic inverse formula for image reconstruction, we developed a GAN based algorithm that provides an efficient mapping between data and image domains. We compare our method against several algorithmic approaches and show that high quality image reconstruction is feasible. Results encourage further research in the application of deep-learning reconstruction techniques in Compton scatter tomography, particularly when inverse reconstruction formulas are unknown. © 2023 IEEE.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Alleviating the Semantic Gap for Generalized fMRI-to-image reconstruction 37

Alleviating the Semantic Gap for Generalized fMRI-to-Image R...

引用

37th conference on Neural Information Processing Systems, NeurIPS 2023

作者： Fang, Tao Zheng, Qian Pan, Gang College of Computer Science and Technology Zhejiang University Hangzhou China The State Key Lab of Brain-Machine Intelligence Zhejiang University Hangzhou China

ISBN: (纸本)9781713899921

Although existing fMRI-to-image reconstruction methods could predict high-quality images, they do not explicitly consider the semantic gap between training and testing data, resulting in reconstruction with unstable and uncertain semantics. This paper addresses the problem of generalized fMRI-to-image reconstruction by explicitly alleviates the semantic gap. Specifically, we leverage the pre-trained CLIP model to map the training data to a compact feature representation, which essentially extends the sparse semantics of training data to dense ones, thus alleviating the semantic gap of the instances nearby known concepts (i.e., inside the training super-classes). Inspired by the robust low-level representation in fMRI data, which could help alleviate the semantic gap for instances that far from the known concepts (i.e., outside the training super-classes), we leverage structural information as a general cue to guide image reconstruction. Further, we quantify the semantic uncertainty based on probability density estimation and achieve Generalized fMRI-to-image reconstruction by adaptively integrating Expanded Semantics and Structural information (GESS) within a diffusion process. Experimental results demonstrate that the proposed GESS model outperforms state-of-the-art methods, and we propose a generalized scenario split strategy to evaluate the advantage of GESS in closing the semantic gap. Our codes are available at https://***/duolala1/GESS. © 2023 Neural information processing systems foundation. All rights reserved.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

CT-Guided, Unsupervised Super-Resolution reconstruction of Single 3D Magnetic Resonance image 26th

CT-Guided, Unsupervised Super-Resolution Reconstruction of S...

引用

26th International conference on Medical image Computing and Computer-Assisted Intervention (MICCAI)

作者： Wang, Jiale Heimann, Alexander F. Tannast, Moritz Zheng, Guoyan Shanghai Jiao Tong Univ Inst Med Robot Sch Biomed Engn 800 Dongchuan Rd Shanghai 200240 Peoples R China Univ Fribourg HFR Cantonal Hosp Dept Orthopaed Surg Fribourg Switzerland

ISBN: (纸本)9783031439063;9783031439070

Deep learning-based algorithms for single MR image (MRI) super-resolution have shown great potential in enhancing the resolution of low-quality images. However, many of these methods rely on supervised training with paired low-resolution (LR) and high-resolution (HR) MR images, which can be difficult to obtain in clinical settings. This is because acquiring HR MR images in clinical settings requires a significant amount of time. In contrast, HR CT images are acquired in clinical routine. In this paper, we propose a CT-guided, unsupervised MRI super-resolution reconstruction method based on joint cross-modality image translation and super-resolution reconstruction, eliminating the requirement of high-resolution MRI for training. The proposed approach is validated on two datasets respectively acquired from two different clinical sites. Well-established metrics including Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Metrics (SSIM), and Learned Perceptual image Patch Similarity (LPIPS) are used to assess the performance of the proposed method. Our method achieved an average PSNR of 32.23, an average SSIM of 0.90 and an average LPIPS of 0.14 when evaluated on data of the first site. An average PSNR of 30.58, an average SSIM of 0.88, and an average LPIPS of 0.10 were achieved by our method when evaluated on data of the second site.

关键词： Unsupervised image super-resolution Cross-modality image translation CT-Guided Magnetic resonance imaging

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：