检索结果-内蒙古大学图书馆

Detail-aware image denoising via structure preserved network and residual diffusion model

VISUAL COMPUTER 2025年第1期41卷 639-658页

作者： Wu, Jing Wu, Hao Yuan, Guowu Yunnan Univ Sch Informat Sci & Engn Kunming Peoples R China

The rapid development of deep learning has led to significant strides in image denoising research and has achieved advanced denoising performance in terms of distortion metrics. However, most denoising models that construct loss functions based on pixel-by-pixel differences cause phenomena, such as blurred edges or over-smoothing in denoised images, unsatisfactory to human perception. Our approach to addressing this issue involves prioritizing visual perceptual quality and efficiently restoring high-frequency details that may have been lost during the point-by-point denoising process, all the while preserving the overall structure of the image. We introduce a structure preserved network to generate cost-effective initial predictions that are subsequently incorporated into a conditional diffusion model as a constraint that closely aligns with the actual images. This allows us to more accurately estimate the distribution of clean images by diffusing from the residuals. We observe that by maintaining image consistency in the initial prediction, we can use a residual diffusion model with lower complexity and fewer iterations to restore the detailed texture for the smoothed parts, ultimately leading to a denoised image sample that is more consistent with the visual perceptual quality. Our method is superior in matching human perceptual metrics, e.g. FID, and maintains its performance even at high noise levels, enabling the preservation of the sharp edge and texture features of the image, while reducing computational costs and equipment requirements. This not only achieves the objective of denoising but also results in enhanced subjective visual effects.

关键词： Image denoising conditional diffusion model Perceptual quality Detailed texture

来源：评论

学校读者我要写书评

暂无评论

Balancing user preferences by social networks: A condition-guided social recommendation model for mitigating popularity bias

引用

NEURAL NETWORKS 2025年 187卷 107317页

作者： He, Xin Fan, Wenqi Wang, Ruobing Wang, Yili Wang, Ying Pan, Shirui Wang, Xin Jilin Univ Sch Artificial Intelligence SAI Dept Comp Sci & Technol 2699 Qianjin St Changchun 130000 Jilin Peoples R China Hong Kong Polytech Univ Kowloon Hong Kong Peoples R China Jilin Univ Coll Comp Sci & Technol Dept Comp Sci & Technol 2699 Qianjin St Changchun 130000 Jilin Peoples R China Griffith Univ Parklands Dr Southport Qld 4222 Australia

Social recommendation models weave social interactions into their design to provide uniquely personalized recommendation results for users. However, social networks not only amplify the popularity bias in recommendation models, resulting in more frequent recommendation of hot items and fewer long-tail items, but also include a substantial amount of redundant information that is essentially meaningless for the model's performance. Existing social recommendation models often integrate the entire social network directly, with little effort to filter or adjust social information to mitigate popularity bias introduced by the social network. In this paper, we propose a Condition-Guided Social Recommendation model (named CGSoRec) to mitigate the model's popularity bias by denoising the social network and adjusting the weights of user's social preferences. More specifically, CGSoRec first includes a Condition-Guided Social Denoising model (CSD) to remove redundant social relations in the social network for capturing users' social preferences with items more precisely. Then, CGSoRec calculates users' social preferences based on denoised social network and adjusts the weights in users' social preferences to make them can counteract the popularity bias present in the recommendation model. At last, CGSoRec includes a Condition-Guided diffusion Recommendation model (CGD) to introduce the adjusted social preferences as conditions to control the recommendation results for a debiased direction. Comprehensive experiments on three real-world datasets demonstrate the effectiveness of our proposed method. The anonymous code is in: https://***/r/CGSoRec-2B72.

关键词： Recommendation system Social recommendation Popularity bias conditional diffusion model

来源：评论

学校读者我要写书评

暂无评论

TSGDiff: Traffic state generative diffusion model using multi-source information fusion

引用

TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES 2025年 174卷

作者： Zhang, Huipeng Dong, Honghui Yang, Zhiqiang Beijing Jiaotong Univ Sch Traff & Transportat 3 Shangyuan Village Beijing 100044 Peoples R China Beijing Jiaotong Univ Beijing Res Ctr Urban Traff Informat Sensing & Ser 3 Shangyuan Village Beijing 100044 Peoples R China

Accurate analysis and prediction of traffic states are fundamental and crucial for intelligent transportation systems, playing a significant role in enhancing the efficiency and safety of traffic systems. Advances in deep learning have promoted the development of traffic prediction. However, some traditional prediction methods primarily rely on historical traffic data to sequentially predict future traffic trends. While some also incorporate one or more influencing factors, such as weather and day of the week, as covariates, they often lack a unified fusion approach to model the impact of these covariates on future traffic states, and they are prone to error accumulation in long-term predictions. To address these challenges, we propose TSGDiff, a novel traffic state generative diffusion model using multi-source information fusion. The proposed method leverages an innovative diffusion model framework and integrates various sources of information, such as traffic data, weather, and weekdays, to enhance the accuracy of traffic state prediction. TSGDiff transforms historical spatiotemporal information and future environment information into feature representations using an attention-based spatiotemporal extraction module and a traffic semantic encoding module, respectively. These feature representations serve as guiding conditions for the diffusion model to generate traffic states. By incorporating the prediction horizon as an input parameter, TSGDiff directly generates future traffic states point-to-point, thereby avoiding error accumulation inherent in iterative prediction methods. To adapt the diffusion model to graph structure road network data, we introduce a Graph Attention U-Net (GAUNet) to capture the spatial correlations in traffic data. Experiments on real-world Beijing traffic datasets demonstrate that TSGDiff significantly outperforms baseline models for long-term predictions and performs comparably for short-term predictions.

关键词： Generative model Traffic state generation Multi-source information conditional diffusion model Speed prediction

来源：评论

学校读者我要写书评

暂无评论

ICDDPM: Image-conditioned denoising diffusion probabilistic model for real-world complex point cloud single view reconstruction

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 259卷

作者： Zhao, Luda Hu, Yihua Yang, Xing Dou, Zhenglei Wu, Qilong Natl Univ Def Technol Coll Elect Engn Hefei Peoples R China Natl Univ Def Technol State Key Lab Pulsed Power Laser Technol Hefei Peoples R China Natl Univ Def Technol Anhui Prov Key Lab Elect Restrict Hefei Peoples R China Hefei Comprehens Natl Sci Ctr Informat Secur Res Ctr Hefei Peoples R China Chinese PLA Anhui Prov Mil Reg Hefei Peoples R China

Point cloud single-view reconstruction (PC-SVR) generates high-quality point clouds from low-cost 2D images, offering a cost-effective solution to the expensive and inefficient acquisition of point cloud for real-world scene typically achieved through expensive LiDAR systems. However, current models that generate point clouds from real-world 2D images (e.g. maritime vessels) still have shortcomings in terms of quality and model generalization. In this paper, we proposed a image-conditioned denoising diffusion probabilistic model (ICDDPM) for real-world complex point cloud single view reconstruction to address these issues. We redesigned the structure of classic diffusion model, use latent shape vectors to seamlessly integrate 2D image encoder, point cloud encoder, and conditional diffusion model, to cater PC-SVR task. By guiding the diffusion process with the 2D images, which serve as crucial conditional information, ICDDPM achieves end-to-end point cloud generation with superior quality. 2D images are also employed as input in the reverse diffusion process to further achieve point cloud generation. We conducted qualitative and quantitative experiments on synthetic dataset ShapeNet and real-world dataset PASCAL3D+ (focused on experiments of vessel point cloud data specifically). The results indicate that the ICDDPM model demonstrates superior performance compared to state-of-the-art models. It is capable of generating point clouds with a greater level of global and local details from various 2D image data. Additionally, the model exhibits strong generalization abilities and requires fewer computational resources.

关键词： Point cloud Single-view reconstruction conditional diffusion model Deep learning

来源：评论

学校读者我要写书评

暂无评论

Neural differentiable modeling with diffusion-based super-resolution for two-dimensional spatiotemporal turbulence

引用

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING 2025年 433卷

作者： Fan, Xiantao Akhare, Deepak Wang, Jian-Xun Univ Notre Dame Dept Aerosp & Mech Engn Notre Dame IN USA Cornell Univ Sibley Sch Mech & Aerosp Engn Ithaca NY 14850 USA

Simulating spatiotemporal turbulence with high fidelity remains a cornerstone challenge in computational fluid dynamics (CFD) due to its intricate multiscale nature and prohibitive computational demands. Traditional approaches typically employ closure models, which attempt to represent small-scale features in an unresolved manner. However, these methods often sacrifice accuracy and lose high-frequency/wavenumber information, especially in scenarios involving complex flow physics. In this paper, we introduce an innovative neural differentiable modeling framework designed to enhance the predictability and efficiency of spatiotemporal turbulence simulations. Our approach features differentiable hybrid modeling techniques that seamlessly integrate deep neural networks with numerical PDE solvers within a differentiable programming framework, synergizing deep learning with physics-based CFD modeling. Specifically, a hybrid differentiable neural solver is constructed on a coarser grid to capture large-scale turbulent phenomena, followed by the application of a Bayesian conditional diffusion model that generates small-scale turbulence conditioned on large-scale flow predictions. Two innovative hybrid architecture designs are studied, and their performance is evaluated through comparative analysis against conventional large eddy simulation techniques with physics-based subgridscale closures and purely data-driven neural solvers. The findings underscore the potential of the neural differentiable modeling framework to significantly enhance the accuracy and computational efficiency of turbulence simulations. This study not only demonstrates the efficacy of merging deep learning with physics-based numerical solvers but also sets a new precedent for advanced CFD modeling techniques, highlighting the transformative impact of differentiable programming in scientific computing.

关键词： Differentiable programming Super-resolution Scientific machine learning Deep neural network conditional diffusion model Generative AI

来源：评论

学校读者我要写书评

暂无评论

Diff-MGR: Dynamic causal graph attention and pattern reproduction guided diffusion model for multivariate time series probabilistic forecasting

引用

INFORMATION SCIENCES 2024年 675卷

作者： Zhao, Tianlong Song, Guangle Li, Xuemei Cui, Lizhen Zhang, Caiming Shandong Univ Sch Software Jinan 250101 Peoples R China Shandong Prov Lab Future Intelligence & Financial Yantai 264005 Peoples R China Philippine Christian Univ Ctr Int Educ Philippine 0900 Netherlands

Time series probability forecasting provides insight into future evolution and its inherent uncertainty from past data. In order to obtain more accurate forecasting as the reliable basis for future decision-making and planning, a new probabilistic prediction model of multivariate time series Diff-MGR is proposed in this study. Diff-MGR innovatively proposes dynamic causal graph attention blocks and pattern reproduction guided prediction blocks to predict the future patterns. Compared with the existing methods, the former captures dynamic unidirectional information flow by using causal graph structure at different periods to provide a real representation of variable relationships. The latter completes a reliable mapping of historical patterns to the future based on the prior knowledge of pattern reproduction. Furthermore, Diff-MGR proposes a novel noise prediction network that can effectively capture dependencies in predicted future patterns and generate probability distributions of future sequences using a conditional diffusion model. Extensive experiments on several real datasets verify the effectiveness of Diff-MGR's components, and show it outperforms existing models in probabilistic forecasting performance.

关键词： Probabilistic forecasting conditional diffusion model Dynamic causal graph attention Pattern reproduction Noise prediction network

来源：评论

学校读者我要写书评

暂无评论

Ghost-UNet: Lightweight model for underwater image enhancement

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年 133卷

作者： Sun, Lingyu Li, Wenqing Xu, Yingjie Hebei Univ Technol Sch Mech Engn Tianjin 300401 Peoples R China

Light scattering and absorption during underwater propagation lead to colour distortion and fogging in underwater images, lacking reference for undegraded images, posing challenges in directly applying the underwater images to real-world scenes. We provide a novel approach for generating underwater images by utilizing a conditional diffusion model to address this issue. Besides, we present a novel approach called Ghost-UNet to address the issue of substantial memory and computational demands in implementations of deep convolutional neural networks. Compared with previous methods, our method exceeds 3.8% and 1.7% in the peak signal-to-noise ratio and underwater colour image quality evaluation respectively. Additionally, our method leads to a 0.016-second gain in test speed. The parameters of Ghost-UNet is 1/39 of a transformer model. Ghost-UNet is highly suitable for real-time underwater image enhancement network implementation.

关键词： conditional diffusion model Underwater image enhancement Generative network Lightweight model

来源：评论

学校读者我要写书评

暂无评论

Object Motion Guided Human Motion Synthesis

引用

ACM TRANSACTIONS ON GRAPHICS 2023年第6期42卷 1-11页

作者： Li, Jiaman Wu, Jiajun Liu, C. Karen Stanford Univ Stanford CA 94305 USA

modeling human behaviors in contextual environments has a wide range of applications in character animation, embodied AI, VR/AR, and robotics. In real-world scenarios, humans frequently interact with the environment and manipulate various objects to complete daily tasks. In this work, we study the problem of full-body human motion synthesis for the manipulation of large-sized objects. We propose Object MOtion guided human MOtion synthesis (OMOMO), a conditional diffusion framework that can generate full-body manipulation behaviors from only the object motion. Since naively applying diffusion models fails to precisely enforce contact constraints between the hands and the object, OMOMO learns two separate denoising processes to first predict hand positions from object motion and subsequently synthesize full-body poses based on the predicted hand positions. By employing the hand positions as an intermediate representation between the two denoising processes, we can explicitly enforce contact constraints, resulting in more physically plausible manipulation motions. With the learned model, we develop a novel system that captures full-body human manipulation motions by simply attaching a smartphone to the object being manipulated. Through extensive experiments, we demonstrate the effectiveness of our proposed pipeline and its ability to generalize to unseen objects. Additionally, as highquality human-object interaction datasets are scarce, we collect a large-scale dataset consisting of 3D object geometry, object motion, and human motion. Our dataset contains human-object interaction motion for 15 objects, with a total duration of approximately 10 hours.

关键词： Human-Object Interaction Animation conditional diffusion model Contact

来源：评论

学校读者我要写书评

暂无评论

PDDM: Poisson-Distribution diffusion model with Multi-Step Weighted Fusion for Medical Image Segmentation

PDDM: Poisson-Distribution Diffusion Model with Multi-Step W...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Chen, Xingyu Chen, Junying South China Univ Technol Sch Software Engn Guangzhou Peoples R China

ISBN: (纸本)9798350359329;9798350359312

Medical image segmentation is challenging because the boundaries of lesions are often hazy or unclear. As a result, how to better distinguish the lesion boundaries to improve segmentation accuracy has become the main focus of research in recent years. The diffusion model, a potent visual image generation model, has extended applications in medical image segmentation. In this work, we propose a Poisson-distribution diffusion model (PDDM), a novel conditional diffusion model based on the Poisson distribution, to fully use the diffusion model's enormous potential in medical image segmentation. To increase the segmentation diversity of diffusion models in discrete spaces and produce more precise segmentation maps, we first explore utilizing Poisson noise as the diffusion kernel rather than Gaussian noise. Then, by making use of the diffusion model's stochastic nature, our model repeatedly samples the original Poisson noise and intermediate latent variables at random to generate a variety of segmentation masks. Moreover, we include a multi-step weighted fusion (MSWF) module in the reasoning process to merge the output of the diffusion model at each step, to improve the robustness of the PDDM's prediction results. Finally, we validate the effectiveness and generalization of our model on three medical image segmentation datasets of different imaging modalities.

关键词： conditional diffusion model Poisson Distribution Multi-Step Weighted Fusion Medical Image Segmentation

来源：评论

学校读者我要写书评

暂无评论

Pre-training Feature Guided diffusion model for Speech Enhancement 25

Pre-training Feature Guided Diffusion Model for Speech Enhan...

引用

25th Interspeech Conference

作者： Yang, Yiyuan Trigoni, Niki Markham, Andrew Univ Oxford Dept Comp Sci Oxford England

Speech enhancement significantly improves the clarity and intelligibility of speech in noisy environments, improving communication and listening experiences. In this paper, we introduce a novel pretraining feature-guided diffusion model tailored for efficient speech enhancement, addressing the limitations of existing discriminative and generative models. By integrating spectral features into a variational autoencoder (VAE) and leveraging pre-trained features for guidance during the reverse process, coupled with the utilization of the deterministic discrete integration method (DDIM) to streamline sampling steps, our model improves efficiency and speech enhancement quality. Demonstrating state-of-the-art results on two public datasets with different SNRs, our model outshines other baselines in efficiency and robustness. The proposed method not only optimizes performance but also enhances practical deployment capabilities, without increasing computational demands.

关键词： speech enhancement denoising conditional diffusion model feature-guided generative model

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：