检索结果-内蒙古大学图书馆

Subnetwork-based federated few-shot semantic segmentation of organ images

Optoelectronics Letters 2025年

作者： Wu Junpeng Zhao Meng Zhang Huanping Key Laboratory of Computer Vision and System of Ministry of Education， School of Computer Science and Engineering， Tianjin University of Technology

Federated learning(FL)， as a distributed learning paradigm， allows multiple medical institutions to collaborate on learning without the need to centralize all client data. However， existing methods pay little attention to more challenging medical image semantic segmentation tasks， especially in the scenario of the imbalanced dataset in federated few-shot learning. In this paper， we propose a subnetwork-based federated few-shot organ image segmentation method. Firstly， individual clients train using local training samples and then upload local model gradients to the server. The server utilizes their respective local model gradients to update the subnetwork maintained on the server and generate aggregation weights for forming personalized model parameters. Through this method， we can learn the similarities between different clients to address data heterogeneity issues. In addition， to enhance the communication efficiency between clients and the server， we have also designed a personalized layer aggregation strategy， which only transmits partial layer model parameters during the communication process to improve communication efficiency. Finally， we conducted experiments on ABD-MRI and ABD-CT datasets to demonstrate the effectiveness of our method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MiT-Unet: Mixed Transformer Unet for Transmission Line Segmentation in UAV Images

引用

IEEE Transactions on Power Delivery 2025年第3期40卷 1279-1288页

作者： Wei, Ning Chen, Jianwei Sun, Shuifa The Hubei Key Laboratory of Intelligent Vision-Based Monitoring for Hydro electric Engineering College of Computer and Information Technology China Three Gorges University Yichang443002 China The Engineering Research Center of Mobile Health Management System Ministry of Education Hangzhou Normal University China

The segmentation of power lines in drone images is one of the challenging tasks in the field of computer vision. Although power lines share the same difficulties with tiny object segmentation, occupying only a very small proportion of pixel areas in the images, the greater challenge is that they also have a very large visual perspective field. Therefore, the results obtained by traditional convolutional neural network-based segmentation methods are still unsatisfactory. To tackle these problems, we propose MiT-Unet (Mixed Transformer Unet), which has an analogous multi-level convolutional neural network structure similar to Unet for encoding and decoding transmission line detailed features. Meanwhile, dealing with excessive scales, we employ an Efficient Self-Attention-based module to enhance the global features of straight lines. Furthermore, in the level transition of this architecture, we propose simple yet efficient upsampling and downsampling modules, TLFM and TLFE, to fuse the global attention feature of power lines. These modules effectively extract and preserve the vulnerable features of power lines during the level transition process. Experimental results demonstrate that our proposed method achieves state-of-the-art performance in transmission line segmentation on the public TTPLA dataset and PLDU dataset. Moreover, the computational efficiency of the proposed model makes it potentially deployable on mobile platforms. The testing code has been released at: https://***/xiachengs/MiT-Unet. ©1986-2012 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

CACP: Covariance-Aware Cross-Domain Prototypes for Domain Adaptive Semantic Segmentation

引用

IEEE Transactions on Multimedia 2025年

作者： Xue, Yanbing Tian, Xinyu Zhang, Feifei Wen, Xianbin Gao, Zan Chen, Shengyong Tianjin University of Technology Key Laboratory of Computer Vision and System Ministry of Education Tianjin300382 China

Domain adaptive semantic segmentation aims to reduce domain shifts / discrepancies between source and target domains, improving the source domain model's generalization ability to the target domain. Recently, prototypical methods, which primarily use single-source or single-target domain prototypes as category centers to aggregate features from both domains, have achieved competitive performance in this task. However, due to large domain shifts, single-source domain prototypes have finite generalization ability and not all source domain knowledge is conducive to model generalization. Single-target domain prototypes are noisy because they are prematurely initialized with all features filtered by pseudo labels, which causes error accumulation in the prototypes. To address these issues, we proposes a covariance-aware cross-domain prototypes method (CACP) to achieve robust domain adaptation. We propose to use both domain prototypes to dynamically rectify pseudo labels in the target domain, effectively reducing the recognition difficulty of hard target domain samples and narrowing the gap between features of the same category in both domains. In addition, to further generalize the model to the target domain, we propose two modules based on covariance correlation, FSPC (Features Selection by Prototypes Covariances) and WSPC (Weighting Source by Prototypes Coefficients), to learn discriminative characteristics. FSPC selects highly correlated features to update target domain prototypes online, denoising and enhancing discriminativeness between categories. WSPC utilizes the correlation coefficients between target domain prototypes and source domain features to weight each point in the source domain, eliminating the information interference from the source domain. In particular, CACP achieves excellent performance on the GTA5 \to Cityscapes and SYNTHIA \to Cityscapes tasks with minimal computational resources and time. © 1999-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Learning to see speckle in the weak laser field through multimode fiber

引用

Optoelectronics Letters 2025年

作者： JI Yunqi SONG Binbin LI Xueqing LI Yonghui The Engineering Research Center of Learning-Based Intelligent System (Ministry of Education)， the Key Laboratory of Computer Vision and Systems (Ministry of Education)， and the School of Computer Science and Engineering， Tianjin University of Technology

Multimode fibers (MMFs) have great potential for endoscopic imaging due to the high number of modes and a small core diameter. Deep learning based on neural networks has received increasing attention in the field of scattering image reconstruction. However， most studies focus on designing complex network architectures to improve reconstruction， but these network models struggle to reconstruct images in a weak laser field. In the paper， a lightweight generative adversarial network model combined with a histogram specification algorithm is designed to reconstruct speckles in the weak laser field through MMF. Experimental results show that the reconstruction results of our algorithm have better metrics. Moreover， the model demonstrates excellent cross-domain generalization ability with regards to the Fashion-MNIST dataset. It is worth mentioning that we found that the speckles after inactivation still retain the ability to be reconstructed， which enhances the robustness of the model

关键词：

来源：评论

学校读者我要写书评

暂无评论

SCSegamba: Lightweight Structure-Aware vision Mamba for Crack Segmentation in Structures

arXiv

引用

arXiv 2025年

作者： Liu, Hui Jia, Chen Shi, Fan Cheng, Xu Chen, Shengyong School of Computer Science and Engineering Tianjin University of Technology China Engineering Research Center of Learning-Based Intelligent System Ministry of Education Key Laboratory of Computer Vision and System Ministry of Education

Pixel-level segmentation of structural cracks across various scenarios remains a considerable challenge. Current methods encounter challenges in effectively modeling crack morphology and texture, facing challenges in balancing segmentation quality with low computational resource usage. To overcome these limitations, we propose a lightweight Structure-Aware vision Mamba Network (SCSegamba), capable of generating high-quality pixel-level segmentation maps by leveraging both the morphological information and texture cues of crack pixels with minimal computational cost. Specifically, we developed a Structure-Aware Visual State Space module (SAVSS), which incorporates a lightweight Gated Bottleneck Convolution (GBC) and a Structure-Aware Scanning Strategy (SASS). The key insight of GBC lies in its effectiveness in modeling the morphological information of cracks, while the SASS enhances the perception of crack topology and texture by strengthening the continuity of semantic information between crack pixels. Experiments on crack benchmark datasets demonstrate that our method outperforms other state-of-the-art (SOTA) methods, achieving the highest performance with only 2.8M parameters. On the multi-scenario dataset, our method reached 0.8390 in F1 score and 0.8479 in mIoU. The code is available at https://***/Karl1109/SCSegamba. © 2025, CC BY.

关键词： State space methods

来源：评论

学校读者我要写书评

暂无评论

Harnessing Light Field Angular Cues and Spatial Geometries for Semantic Segmentation

Harnessing Light Field Angular Cues and Spatial Geometries f...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Chen Jia Fan Shi Xu Cheng School of Computer Science and Engineering The Engineering Research Center of Learning-Based Intelligent System (Ministry of Education) The Key Laboratory of Computer Vision and System (Ministry of Education) Tianjin University of Technology Tianjin China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

4D light field imaging captures rich spatial-angular information, providing essential geometric cues for semantic segmentation tasks. In this paper, we introduce a novel backbone network called the Light Field Extraction Interaction Network (LFEI-Net). LFEI-Net excels in extracting global structures and multi-scale spatial-angular features, capturing feature dependencies through channel modeling and diverse feature interactions. Unlike traditional methods that depend on pyramid and dilated feature extraction, LFEI-Net pioneers an efficient method by integrating large-scale horizontal depth-wise convolution (HDWC) and vertical depth-wise convolution (VDWC) with interactive operations for comprehensive spatial multi-scale feature extraction. Furthermore, we present the Multi-Angular Modeling (MAM) module, which effectively captures scene angle variations from multiple perspectives and precisely delineates object boundaries, thereby improving model adaptability. Our experimental evaluations on two datasets demonstrate that LFEI-Net significantly outperforms state-ofthe-art (SOTA) 2D and 4D light field semantic segmentation methods, achieving mean Intersection over Union (mIoU) of 83.72% and 86.88%, respectively.

关键词： Geometry Adaptation models Convolution Semantic segmentation Imaging Feature extraction Light fields Acoustics Speech processing

来源：评论

学校读者我要写书评

暂无评论

Prior Knowledge-Driven Hybrid Prompter Learning for RGB-Event Tracking

引用

IEEE Transactions on Circuits and systems for Video Technology 2025年

作者： Wang, Mianzhao Shi, Fan Cheng, Xu Chen, Shengyong Tianjin University of Technology Engineering Research Center of Learning-Based Intelligent System Ministry of Education Key Laboratory of Computer Vision and System School of Computer Science and Engineering Tianjin300384 China Technical University of Denmark Department of Technology Management and Economics Kongens Lyngby Denmark

Event data can asynchronously capture variations in light intensity, thereby implicitly providing valuable complementary cues for RGB-Event tracking. Existing methods typically employ a direct interaction mechanism to fuse RGB and event data. However, due to differences in imaging mechanisms, the representational disparity between these two data types is not fixed, which can lead to tracking failures in certain challenging scenarios. To address this issue, we propose a novel prior knowledge-driven hybrid prompter learning framework for RGB-Event tracking. Specifically, we develop a frame-event hybrid prompter that leverages prior tracking knowledge from the foundation model as intermediate modal support to mitigate the heterogeneity between RGB and event data. By leveraging its rich prior tracking knowledge, the intermediate modal reduces the gap between the dense RGB and sparse event data interactions, effectively guiding complementary learning between modalities. Meanwhile, to mitigate the internal learning disparities between the lightweight hybrid prompter and the deep transformer model, we introduce a pseudo-prompt learning strategy that lies between full fine-tuning and partial fine-tuning. This strategy adopts a divide-and-conquer approach to assign different learning rates to modules with distinct functions, effectively reducing the dominant influence of RGB information in complex scenarios. Extensive experiments conducted on two public RGB-Event tracking datasets show that the proposed HPL outperforms state-of-the-art tracking methods, achieving exceptional performance. © 1991-2012 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

An Encoder-Decoder Model Based On Spiking Neural Networks For Address Event Representation Object Recognition

引用

IEEE Transactions on Cognitive and Developmental systems 2025年

作者： Du, Sichun Zhu, Haodi Zhang, Yang Hong, Qinghui Hunan University College of Computer Science and Electronic Engineering Changsha418002 China Shenzhen University Computer Vision Institute School of Computer Science and Software Engineering National Engineering Laboratory for Big Data System Computing Technology Guangdong Key Laboratory of Intelligent Information Processing Shenzhen518060 China

Address event representation (AER) object recognition task has attracted extensive attention in neuromorphic vision processing. The spike-based and event-driven computation inherent in the spiking neural network (SNN) provides an energy-saving solution for AER object recognition. However, SNN with spike timing dependent plasticity (STDP) learning rule has not achieved satisfying AER object recognition performance. This work proposes an SNN-based encoder-decoder model to improve the recognition performance of AER objects. An STDP-based locally connected spiking neural network (LC-SNN) is proposed as an encoder to extract rich spatiotemporal features from AER event flows more flexibly. After the encoder extracts and learns primary features, we propose a fully connected spiking neural network (FC-SNN) based on the reward-modulated spike-timing-dependent plasticity (R-STDP) learning rule as a decoder to learn higher-level features for classification. In addition, we improved the winner-take-all (WTA) mechanisms and R-STDP learning rule in the decoder based on the reward and punish decision, enabling the network to perform better. The experiments are performed on the N-MNIST, MNIST-DVS, and the DVS Gesture datasets, improving the accuracy of the best existing plasticity-based SNN by 0.19%, 0.27% and 1.35%, respectively. © 2016 IEEE.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis

DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Kaijun Deng Dezhi Zheng Jindong Xie Jinbao Wang Weicheng Xie Linlin Shen Siyang Song Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Guangdong Provincial Key Laboratory of Intelligent Information Processing Department of Computer Science University of Exeter

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Accurately synthesizing talking face videos and capturing fine facial features for individuals with long hair presents a significant challenge. To tackle these challenges in existing methods, we propose a decomposed per-embedding Gaussian fields (DEGSTalk), a 3D Gaussian Splatting (3DGS)-based talking face synthesis method for generating realistic talking faces with long hairs. Our DEGSTalk employs Deformable Pre-Embedding Gaussian Fields, which dynamically adjust pre-embedding Gaussian primitives using implicit expression coefficients. This enables precise capture of dynamic facial regions and subtle expressions. Additionally, we propose a Dynamic Hair-Preserving Portrait Rendering technique to enhance the realism of long hair motions in the synthesized videos. Results show that DEGSTalk achieves improved realism and synthesis quality compared to existing approaches, particularly in handling complex facial dynamics and hair preservation. Our code is available at https://***/CVI-SZU/DEGSTalk.

关键词： Hair Training Three-dimensional displays Dynamics Signal processing Rendering (computer graphics) Noise measurement Speech processing Faces Videos

来源：评论

学校读者我要写书评

暂无评论

Dual Encoders for Diffusion-based Image Inpainting

Dual Encoders for Diffusion-based Image Inpainting

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Dezhi Zheng Kaijun Deng Jinbao Wang Linlin Shen Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Current diffusion-based inpainting models struggle to preserve unmasked regions or generate highly coherent content. Additionally, it is hard for them to generate meaningful content for 3D inpainting. To tackle these challenges, we design a plug-and-play branch that runs through the entire generation process to enhance existing models. Specifically, we utilize dual encoders - a Convolutional Neural Network (CNN) encoder and the pre-trained Variational AutoEncoder (VAE) encoder, to encode masked images. The latent code and the feature map from the dual encoders are fed to diffusion models simultaneously. In addition, we apply Zero-padded initialization to solve the problem of mode collapse caused by this branch. Experiments on BrushBench and EditBench demonstrate that models with our plug-and-play branch can improve the coherence of inpainting, and our model achieves new state-of-the-art results.

关键词： Visualization Three-dimensional displays Codes Autoencoders Coherence Signal processing Diffusion models Decoding Convolutional neural networks Speech processing

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：