Pattern Recognition Letters最新文献_第8页

Domain generalization using action sequences for egocentric action recognition 基于动作序列的自中心动作识别领域泛化

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-24 DOI: 10.1016/j.patrec.2025.06.010

Amirshayan Nasirimajd, Chiara Plizzari, Simone Alberto Peirone, Marco Ciccone, Giuseppe Averta, Barbara Caputo

{"title":"Domain generalization using action sequences for egocentric action recognition","authors":"Amirshayan Nasirimajd, Chiara Plizzari, Simone Alberto Peirone, Marco Ciccone, Giuseppe Averta, Barbara Caputo","doi":"10.1016/j.patrec.2025.06.010","DOIUrl":"10.1016/j.patrec.2025.06.010","url":null,"abstract":"<div><div>Recognizing human activities from visual inputs, particularly through a first-person viewpoint, is essential for enabling robots to replicate human behavior. Egocentric vision, characterized by cameras worn by observers, captures diverse changes in illumination, viewpoint, and environment. This variability leads to a notable drop in the performance of Egocentric Action Recognition models when tested in environments not seen during training. In this paper, we tackle these challenges by proposing a domain generalization approach for Egocentric Action Recognition. Our insight is that action sequences often reflect consistent user intent across visual domains. By leveraging <em>action sequences</em>, we aim to enhance the model’s generalization ability across unseen environments. Our proposed method, named SeqDG, introduces a visual-text sequence reconstruction objective (SeqRec) that uses contextual cues from both text and visual inputs to reconstruct the central action of the sequence. Additionally, we enhance the model’s robustness by training it on mixed sequences of actions from different domains (SeqMix). We validate SeqDG on the EGTEA and EPIC-KITCHENS-100 datasets. Results on EPIC-KITCHENS-100, show that SeqDG leads to +2.4% relative average improvement in cross-domain action recognition in unseen environments, and on EGTEA the model achieved +0.6% Top-1 accuracy over SOTA in intra-domain action recognition. Code and Data: <span><span>Github.com/Ashayan97/SeqDG</span><svg><path></path></svg></span></div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 213-220"},"PeriodicalIF":3.9,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144491068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

RESAMPL-UDA: Leveraging foundation models for unsupervised domain adaptation in biomedical images resamp - uda：利用基础模型在生物医学图像中进行无监督域适应

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-24 DOI: 10.1016/j.patrec.2025.06.007

Alexandre Stenger , Etienne Baudrier , Benoît Naegel , Nicolas Passat

{"title":"RESAMPL-UDA: Leveraging foundation models for unsupervised domain adaptation in biomedical images","authors":"Alexandre Stenger , Etienne Baudrier , Benoît Naegel , Nicolas Passat","doi":"10.1016/j.patrec.2025.06.007","DOIUrl":"10.1016/j.patrec.2025.06.007","url":null,"abstract":"<div><div>Large annotated datasets and new models have led to significant improvements in supervised semantic segmentation. On the other side, Unsupervised Domain Adaptation (UDA) for Semantic Segmentation is still an arduous open research topic. While new ideas frequently come out based on recent findings, best methods still rely on basic techniques such as the use of pseudo-labels on target for self-training. Nonetheless, such methods fail when applied to difficult UDA cases like Biomedical Images where the domain shift is too high, leading to pseudo-labels of poor quality. In this work, we propose RESAMPL-UDA (<strong>RE</strong>fined <strong>SAM</strong>-based <strong>P</strong>seudo <strong>L</strong>abel <strong>UDA</strong>), an unsupervised domain adaptation method that effectively integrates zero-shot predictions from the Segment Anything (SAM) model. Given the high complexity and variability of biomedical images, SAM alone often produces detailed segmentations without necessarily capturing the intended structures. To address this, our method involves training a dedicated refinement network on source domain data to selectively enhance SAM-generated masks. These refined segmentations then serve as reliable pseudo-labels within our UDA framework, significantly facilitating the adaptation process. Experiments on 8 adaptation cases demonstrate that our method outperforms the state of the art. In addition, we extend successfully our work to Source-Free Unsupervised Domain Adaptation, demonstrating its versatility. The code is available : <span><span>https://github.com/alex-stenger/RESAMPL-UDA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 221-227"},"PeriodicalIF":3.9,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spectral ensemble clustering from graph reconstruction with auto-weighted cluster 基于自加权聚类的图重构光谱系综聚类

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-23 DOI: 10.1016/j.patrec.2025.05.025

Xiaojun Yang , Zhenhao Zheng , Jieming Xie , Weihao Zhao , JingJing Xue , Feiping Nie

{"title":"Spectral ensemble clustering from graph reconstruction with auto-weighted cluster","authors":"Xiaojun Yang , Zhenhao Zheng , Jieming Xie , Weihao Zhao , JingJing Xue , Feiping Nie","doi":"10.1016/j.patrec.2025.05.025","DOIUrl":"10.1016/j.patrec.2025.05.025","url":null,"abstract":"<div><div>Ensemble clustering (EC) integrates the base clusterings into a consensus result, which is more robust and effective. A common approach involves constructing a co-association (CA) matrix, and then some individual algorithms are applied to it. However, this approach does not account for the variations in quality across different base clusterings and clusters. Although some weighted ensemble clustering approaches have been introduced, their assigned weights remain fixed once they are determined according to specific principles. Additionally, this method isolates the process of building the co-association matrix from the generation of clustering outcomes. To address these issues, we propose a novel clustering algorithm that adaptively adjust cluster weights. In our method, we define a novel weighted CA matrix and apply a self-weighting framework to automatically assign weights to clusters. Then, the construction of the consensus graph and spectral clustering are integrated into a single framework. Finally, an effective optimization algorithm, the coordinate descent method, is used to directly produce a discrete label matrix. The effectiveness of the proposed approach is validated through experiments on both synthetic and real-world datasets.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 243-249"},"PeriodicalIF":3.9,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144518471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spike-TBR: A noise resilient neuromorphic event representation 刺突- tbr：噪声弹性神经形态事件表征

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-23 DOI: 10.1016/j.patrec.2025.05.018

Gabriele Magrini , Federico Becattini , Luca Cultrera , Lorenzo Berlincioni , Pietro Pala , Alberto Del Bimbo

{"title":"Spike-TBR: A noise resilient neuromorphic event representation","authors":"Gabriele Magrini , Federico Becattini , Luca Cultrera , Lorenzo Berlincioni , Pietro Pala , Alberto Del Bimbo","doi":"10.1016/j.patrec.2025.05.018","DOIUrl":"10.1016/j.patrec.2025.05.018","url":null,"abstract":"<div><div>Event cameras offer significant advantages over traditional frame-based sensors, including higher temporal resolution, lower latency and dynamic range. However, efficiently converting event streams into formats compatible with standard computer vision pipelines remains a challenging problem, particularly in the presence of noise. In this paper, we propose Spike-TBR, a novel event-based encoding strategy based on Temporal Binary Representation (TBR), addressing its vulnerability to noise by integrating spiking neurons. Spike-TBR combines the frame-based advantages of TBR with the noise-filtering capabilities of spiking neural networks, creating a more robust representation of event streams. We evaluate four variants of Spike-TBR, each using different spiking neurons, across multiple datasets, demonstrating superior performance in noise-affected scenarios while improving the results on clean data. Our method bridges the gap between spike-based and frame-based processing, offering a simple noise-resilient solution for event-driven vision applications.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 198-205"},"PeriodicalIF":3.9,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144491254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Segment Anything Model for detecting salient objects with accurate prompting and Ladder Directional Perception 用精确提示和阶梯方向感知来检测显著物体的分割模型

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-21 DOI: 10.1016/j.patrec.2025.06.002

Yuze Sun, Hongwei Zhao, Jianhang Zhou

引用次数: 0

United diverse subgraph for graph incremental learning 图增量学习的统一多元子图

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-21 DOI: 10.1016/j.patrec.2025.06.004

Yanfeng Sun, Jiaxing Zhang, Qi Zhang, Shaofan Wang, Baocai Yin

{"title":"United diverse subgraph for graph incremental learning","authors":"Yanfeng Sun, Jiaxing Zhang, Qi Zhang, Shaofan Wang, Baocai Yin","doi":"10.1016/j.patrec.2025.06.004","DOIUrl":"10.1016/j.patrec.2025.06.004","url":null,"abstract":"<div><div>Graph incremental learning has emerged as a powerful graph deep learning framework, showcasing superior performance in addressing the evolving nature of graph data. However, catastrophic forgetting, which involves forgetting previously learned knowledge and overfitting to new data for sequential graph learning tasks, has become one of the most crucial challenges for graph incremental learning. Recent research has highlighted the significance of experience replay for effective anti-forgetting. This paper proposes a novel graph incremental learning model based on United Diverse Subgraph (UDS) for experience replay. This model firstly samples diverse nodes based on the uncertainty of each node for current task. Consequently, the unsampled nodes are pooled together into a supernode to extract global features from the unsampled nodes. Moreover, the structural relationships between nodes are established to form the final diverse subgraph for experience replay. This approach can capture both rich local and global information from current graph, which significantly reduces the space complexity of storing subgraphs. Extensive experiments conducted on various graph incremental learning datasets, consistently demonstrate the superior performance of our approach compared to existing graph incremental learning in the context of node classification.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 206-212"},"PeriodicalIF":3.9,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144491255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Knowledge-sharing hierarchical memory fusion network for scribble-supervised video salient object detection 基于知识共享层次记忆融合网络的涂鸦监督视频显著目标检测

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-21 DOI: 10.1016/j.patrec.2025.06.003

Tao Jiang , Feng Hou , Yi Wang , Guangzhu Chen , Ruili Wang

{"title":"Knowledge-sharing hierarchical memory fusion network for scribble-supervised video salient object detection","authors":"Tao Jiang , Feng Hou , Yi Wang , Guangzhu Chen , Ruili Wang","doi":"10.1016/j.patrec.2025.06.003","DOIUrl":"10.1016/j.patrec.2025.06.003","url":null,"abstract":"<div><div>Scribble annotations offer a practical alternative to pixel-wise labels in video salient object detection (V-SOD). However, their sparse foreground coverage and ambiguous boundaries introduce background interference and error propagation, degrading segmentation accuracy across frames. To address this issue, we propose a novel Knowledge-sharing Hierarchical Memory Fusion Network (KHMF-Net) for scribble-supervised V-SOD. The core of our framework is a Hierarchical Memory Bank (HMB) that stores initial scribbles, historical high-confidence regions, and historical full salient maps, enabling long-term spatiotemporal context modeling to suppress error propagation. Additionally, we introduce an Adaptive Memory Fusion (AMF) module to dynamically integrate multi-confidence features, providing reliable guidance during salient mask expansion. To address background interference, we design an Interactive Equalized Matching (IEM) module with reference-wise softmax, ensuring balanced contributions from reference frame pixels. A dual-attention knowledge-sharing mechanism is further proposed to enhance IEM by transferring high-performance attention features from a Teacher to a Student module, improving segmentation accuracy. Experimental results demonstrate that KHMF-Net’s hierarchical memory architecture and effective background-target discrimination enable state-of-the-art performance on three scribble-annotated datasets, even exceeding some fully supervised approaches. The module and predicted maps are publicly available at <span><span>https://github.com/TOMMYWHY/KHMF-Net</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 177-183"},"PeriodicalIF":3.9,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144471983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

F-MMD-DBA: Frobenius-norm Maximum Mean Discrepancy for domain bi-classifier adversarial 领域双分类器对抗的frobenius -范数最大平均差异

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-21 DOI: 10.1016/j.patrec.2025.06.005

Zichao Cai , Zongze Wu , Yanyun Qu , Deyu Zeng

{"title":"F-MMD-DBA: Frobenius-norm Maximum Mean Discrepancy for domain bi-classifier adversarial","authors":"Zichao Cai , Zongze Wu , Yanyun Qu , Deyu Zeng","doi":"10.1016/j.patrec.2025.06.005","DOIUrl":"10.1016/j.patrec.2025.06.005","url":null,"abstract":"<div><div>Unsupervised domain bi-classifier adversarial approaches are promising methods to deal with domain shifts. Combined with metric learning, they significantly reduce ambiguous predictions. However, they have challenges in the leverage of global information, the relationships between subdomains, the consistency of feature representation, and the conflict due to the bi-classifier adversarial paradigm and usage of local information. Here, we propose Frobenius-norm Maximum Mean Discrepancy for Domain Bi-classifier Adversarial (F-MMD-DBA) to alleviate these by imposing and integrating global and local constraints. Maximum Mean Discrepancy, as the global constraint, can utilize global information. As the local constraint, Frobenius-norm focuses on the relationship of subdomains and avoids conflict. The integration of global and local constraints increases the consistency of feature representation. Ours has greatly improved the accuracy of digits classification and object recognition tasks. The code will be available at <span><span>https://github.com/JackyDeepLearning/F-MMD-DBA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 191-197"},"PeriodicalIF":3.9,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144491253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multimodal, context-based dataset of children with Post Traumatic Stress Disorder 创伤后应激障碍儿童的多模式、基于情境的数据集

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-20 DOI: 10.1016/j.patrec.2025.05.003

Saandeep Aathreya , Tara Nourivandi , Alison Salloum , Leigh J. Ruth , Eric A. Storch , Shaun Canavan

{"title":"Multimodal, context-based dataset of children with Post Traumatic Stress Disorder","authors":"Saandeep Aathreya , Tara Nourivandi , Alison Salloum , Leigh J. Ruth , Eric A. Storch , Shaun Canavan","doi":"10.1016/j.patrec.2025.05.003","DOIUrl":"10.1016/j.patrec.2025.05.003","url":null,"abstract":"<div><div>The conventional method of diagnosing Post Traumatic Stress Disorder by a clinician has been subjective in nature by taking specific events/context in consideration. Developing AI-based solutions to these sensitive areas calls for adopting similar methodologies. Considering this, we propose a de-identified dataset of children subjects who are clinically diagnosed with/without PTSD in multiple contexts. This datset can help facilitate future research in this area. For each subject, in the dataset, the participant undergoes several sessions with clinicians and/or guardian that brings out various emotional response from the participant. We collect videos of these sessions and for each video, we extract several facial features that detach the identity information of the subjects. These include facial landmarks, head pose, action units (AU), and eye gaze. To evaluate this dataset, we propose a baseline approach to identifying PTSD using the encoded action unit (AU) intensities of the video frames as the features. We show that AU intensities intrinsically captures the expressiveness of the subject and can be leveraged in modeling PTSD solutions. The AU features are used to train a transformer for classification where we propose encoding the low-dimensional AU intensity vectors using a learnable Fourier representation. We show that this encoding, combined with a standard Multilayer Perceptron (MLP) mapping of AU intensities yields a superior result when compared to its individual counterparts. We apply the approach to various contexts of PTSD discussions (e.g., Clinician-child discussion) and our experiments show that using context is essential in classifying videos of children.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 228-235"},"PeriodicalIF":3.9,"publicationDate":"2025-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Zero-X21: Scale-agnostic image feature conditioned INR for multi-modal and multi-planar anisotropic MRI inter-slice interpolation Zero-X21：用于多模态和多平面各向异性MRI层间插值的尺度不可知图像特征条件INR

IF 3.9 3区计算机科学

Pattern Recognition Letters Pub Date : 2025-06-19 DOI: 10.1016/j.patrec.2025.06.001

Zibo Ma , Jianfei Huo , Guanchun Yin , Bo Zhang , Xiuzhuang Zhou , Zhen Cui , Wendong Wang

{"title":"Zero-X21: Scale-agnostic image feature conditioned INR for multi-modal and multi-planar anisotropic MRI inter-slice interpolation","authors":"Zibo Ma , Jianfei Huo , Guanchun Yin , Bo Zhang , Xiuzhuang Zhou , Zhen Cui , Wendong Wang","doi":"10.1016/j.patrec.2025.06.001","DOIUrl":"10.1016/j.patrec.2025.06.001","url":null,"abstract":"<div><div>Combining different planes of multi-contrast anisotropic Magnetic Resonance Imaging (MRI) into high-resolution isotropic and multi-contrast MRI can provide richer diagnostic information. However, this is challenging due to the inherent inconsistencies in image features and other aspects across different views. This challenge can be referred to as the inter-slice interpolation problem of multi-modal and multi-planar anisotropic MRI.</div><div>Currently, Implicit Neural Representations (INR) offer the advantage of handling arbitrary up-sampling scales (integer or fractional). However, existing INR-based super-resolution methods often suffer from limitations, including poor generalization ability, inadequate multi-modal information interaction, and limited capacity for image feature modulation.</div><div>To address these limitations, we propose Zero-X21, a novel image feature-conditioned INR framework specifically designed for the inter-slice interpolation problem of anisotropic MRI. Leveraging the inherent continuity of INRs, the Zero-X21 framework excels in achieving high-quality results across arbitrary up-sampling scales, surpassing other volumetric super-resolution methods.</div><div>Experimental results on a brain MRI dataset demonstrate that the Zero-X21 framework achieves state-of-the-art performance for inter-slice interpolation. Notably, a single trained Zero-X21 model can effectively handle arbitrary up-sampling scales, making it a versatile and efficient solution for this challenging task.</div></div>","PeriodicalId":54638,"journal":{"name":"Pattern Recognition Letters","volume":"196 ","pages":"Pages 236-242"},"PeriodicalIF":3.9,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144518465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0