Pattern Recognition最新文献

Trace Back and Go Ahead: Completing partial annotation for continual semantic segmentation

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-02 DOI: 10.1016/j.patcog.2025.111613

Yuxuan Luo , Jinpeng Chen , Runmin Cong , Horace Ho Shing Ip , Sam Kwong

{"title":"Trace Back and Go Ahead: Completing partial annotation for continual semantic segmentation","authors":"Yuxuan Luo , Jinpeng Chen , Runmin Cong , Horace Ho Shing Ip , Sam Kwong","doi":"10.1016/j.patcog.2025.111613","DOIUrl":"10.1016/j.patcog.2025.111613","url":null,"abstract":"<div><div>Existing Continual Semantic Segmentation (CSS) methods effectively address the issue of <em>background shift</em> in regular training samples. However, this issue persists in exemplars, <em>i.e.</em>, replay samples, which is often overlooked. Each exemplar is annotated only with the classes from its originating task, while other past classes and the current classes during replay are labeled as <em>background</em>. This partial annotation can erase the network’s knowledge of previous classes and impede the learning of new classes. To resolve this, we introduce a new method named Trace Back and Go Ahead (TAGA), which utilizes a backward annotator model and a forward annotator model to generate pseudo-labels for both regular training samples and exemplars, aiming at reducing the adverse effects of incomplete annotations. This approach effectively mitigates the risk of incorrect guidance from both sample types, offering a comprehensive solution to <em>background shift</em>. Additionally, due to a significantly smaller number of exemplars compared to regular training samples, the class distribution in the sample pool of each incremental task exhibits a long-tailed pattern, potentially biasing classification towards incremental classes. Consequently, TAGA incorporates a class-equilibrium sampling strategy that adaptively adjusts the sampling frequencies based on the ratios of exemplars to regular samples and past to new classes, counteracting the skewed distribution. Extensive experiments on two public datasets, Pascal VOC 2012 and ADE20K, demonstrate that our method surpasses state-of-the-art methods.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111613"},"PeriodicalIF":7.5,"publicationDate":"2025-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768429","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Hyperspectral image restoration via the collaboration of low-rank tensor denoising and completion

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-01 DOI: 10.1016/j.patcog.2025.111629

Tianheng Zhang , Jianli Zhao , Sheng Fang , Zhe Li , Qing Zhang , Maoguo Gong

{"title":"Hyperspectral image restoration via the collaboration of low-rank tensor denoising and completion","authors":"Tianheng Zhang , Jianli Zhao , Sheng Fang , Zhe Li , Qing Zhang , Maoguo Gong","doi":"10.1016/j.patcog.2025.111629","DOIUrl":"10.1016/j.patcog.2025.111629","url":null,"abstract":"<div><div>Hyperspectral images (HSIs) are always damaged by various types of noise during acquisition and transmission. Low-rank tensor denoising methods have achieved state-of-the-art results in current HSIs restoration tasks. However, all these methods remove the mixed noise in HSI based on the representation of image prior information. In this paper, we consider a problem for the first time: Structured noise like stripes and deadlines confounds image priors, hindering effective image-noise separation in current approaches. Motivated by this, a new HSI restoration model based on the collaboration of low-rank tensor denoising and completion (LR-TDTC) is proposed. Firstly, the structured noise detection algorithm is applied to identify the positions of structured noise such as stripes and deadlines, achieving the separation of unstructured noise and structured noise. The entries in the structured noisy area are removed. Then, for unstructured noise, a tensor denoising module (TD) based on image prior representation is introduced to separate images and noise. For structured noise, a tensor completion module (TC) based on full-mode-augmentation tensor train rank minimization is introduced to complete the noise area. Finally, the two modules collaborate through the mutual utilization of information to achieve the restoration of the entire image. To solve the LR-TDTC model, a variable tessellation iterative algorithm (VTI) is proposed. VTI utilizes a serialization strategy to enable TD and TC modules to effectively utilize each other's latest iteration results, achieving efficient collaboration between the two. The mixed noise removal experiments on multiple HSIs show that the proposed method has outstanding advantages.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111629"},"PeriodicalIF":7.5,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Deepfake detection with domain generalization and mask-guided supervision

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-31 DOI: 10.1016/j.patcog.2025.111622

Jicheng Li , Yongjian Hu , Beibei Liu , Huimin She , Chang-Tsun Li

{"title":"Deepfake detection with domain generalization and mask-guided supervision","authors":"Jicheng Li , Yongjian Hu , Beibei Liu , Huimin She , Chang-Tsun Li","doi":"10.1016/j.patcog.2025.111622","DOIUrl":"10.1016/j.patcog.2025.111622","url":null,"abstract":"<div><div>Most existing deepfake (video face forgery) detectors work well in intra-dataset testing, but their performance degrades severely in cross-dataset testing. Cross-dataset generalization remains a major challenge. Since domain generalization (DG) aims to learn domain-invariant features while suppressing domain specific features, we propose a DG framework for improving face forgery detection in this study. Our detector consists of two modules. The first module learns both spatial and spectral features from frame images. The second one learns high-level feature patterns from the outputs of the first module, and constructs the classification features with the help of face mask-guided supervision. The classification result is fine-tuned by a confidence-based correction mechanism. The DG framework is realized through a bi-level optimization process. Extensive experiments demonstrate that our detector works effectively in both intra- and cross-dataset testing. Compared with 8 typical methods, it has the best overall performance and the highest robustness against common perturbations.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111622"},"PeriodicalIF":7.5,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Single source domain generalization for palm biometrics

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-29 DOI: 10.1016/j.patcog.2025.111620

Congcong Jia , Xingbo Dong , Yen Lung Lai , Andrew Beng Jin Teoh , Ziyuan Yang , Xiaoyan Zhang , Liwen Wang , Zhe Jin , Lianqiang Yang

引用次数: 0

A novel heterogeneous data classification approach combining gradient boosting decision trees and hybrid structure model

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-28 DOI: 10.1016/j.patcog.2025.111614

Feng Xu , Yuting Huang , Hui Wang , Zizhu Fan

{"title":"A novel heterogeneous data classification approach combining gradient boosting decision trees and hybrid structure model","authors":"Feng Xu , Yuting Huang , Hui Wang , Zizhu Fan","doi":"10.1016/j.patcog.2025.111614","DOIUrl":"10.1016/j.patcog.2025.111614","url":null,"abstract":"<div><div>Graph neural network (GNN) is crucial in graph representation learning tasks. However, when the feature of graph network nodes is complex, such as those originating from heterogeneous data or multi-view data, graph neural network methods encounter difficulties. It is well known that gradient boosting decision trees (GBDT) excel at handling heterogeneous tabular data, while GNN and HGNN perform well with low-order and high-order sparse matrices. Therefore, we propose a method that combines their strengths by using GBDT to handle heterogeneous features, while a hybrid structured model (HSM) based on GNN and hypergraph neural network (HGNN), which can effectively capture both low-order and high-order information, backpropagates gradients to the GBDT. The proposed GBDT-HSM algorithm performs well on four structured tabular datasets and two multi-view datasets. It achieves state-of-the-art performance, showcasing its potential in addressing the challenges of heterogeneous data classification. The code is available at <span><span>https://github.com/zzfan3/GBDT-HSM</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111614"},"PeriodicalIF":7.5,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143746200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Adversarial temporal sentence grounding by learning from external data

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-28 DOI: 10.1016/j.patcog.2025.111621

Tingting Han , Kai Wang , Jun Yu , Sicheng Zhao , Jianping Fan

{"title":"Adversarial temporal sentence grounding by learning from external data","authors":"Tingting Han , Kai Wang , Jun Yu , Sicheng Zhao , Jianping Fan","doi":"10.1016/j.patcog.2025.111621","DOIUrl":"10.1016/j.patcog.2025.111621","url":null,"abstract":"<div><div>Temporal sentence grounding (TSG) aims to localize the temporal moment that semantically corresponds to a given natural language query in the untrimmed video. Great efforts have been made to solve the problem in both fully supervised and weakly supervised settings. However, fully supervised methods heavily rely on manually annotated start and end timestamps which are arduous to obtain, while weakly supervised methods suffer from performance issues due to the lack of supervision. In this paper, we propose to solve the temporal sentence grounding by exploring external data. Specifically, we design an Adversarial Temporal Sentence Grounding (ATSG) framework, comprising a proposal generator and a semantic discriminator which is firstly pre-trained on external data. Benefiting from the pre-training, the semantic discriminator possesses the ability to distinguish cross-modal semantic similarities and encourages the proposal generator to produce more accurate candidates. In addition, we use an adversarial training process in the joint optimization stage where the proposal generator and the semantic discriminator compete alternately, ultimately leading to improved TSG performance. We conduct extensive experiments on two public benchmarks, i.e., ActivityNet Captions and Charades-STA, and the results demonstrate that the proposed ATSG network achieves state-of-the-art performance.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111621"},"PeriodicalIF":7.5,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Leveraging multi-level regularization for efficient Domain Adaptation of Black-box Predictors

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-28 DOI: 10.1016/j.patcog.2025.111611

Wei Li , Wenyi Zhao , Xipeng Pan , Pengcheng Zhou , Huihua Yang

{"title":"Leveraging multi-level regularization for efficient Domain Adaptation of Black-box Predictors","authors":"Wei Li , Wenyi Zhao , Xipeng Pan , Pengcheng Zhou , Huihua Yang","doi":"10.1016/j.patcog.2025.111611","DOIUrl":"10.1016/j.patcog.2025.111611","url":null,"abstract":"<div><div>Source-free domain adaptation (SFDA) aims to adapt a source-trained model to a target domain without exposing the source data, addressing concerns about data privacy and security. Nevertheless, this paradigm is still susceptible to data leakage due to potential adversarial attacks on the source model. Domain adaptation of black-box predictors (DABP) offers an alternative approach that does not require access to both the source-domain data and the predictor parameters. Existing DABP methods, however, have several significant drawbacks: (1) Lightweight models may underperform due to limited learning capacity. (2) The potential of the target data is not fully harness to learn the structure of the target domain. (3) Focusing exclusively on input-level or network-level regularization renders feature representations susceptible to noisy pseudo labels, degrading performance. Aiming at these limitations, we introduce a novel approach referred to as <strong>M</strong>ulti-<strong>L</strong>evel <strong>R</strong>egularization (MLR) for efficient black-box domain adaptation from network-level, input-level, and feature-level. Our MLR framework comprises a teacher–student network that allows peer networks to utilize pseudo labels generated by each other for supplementary guidance, thereby learning diverse target representations and alleviating overfitting on the source domain. At the input level, we integrate both local and global interpolation consistency training strategies to capture the inherent structure of the target data. Furthermore, by leveraging input-level and network-level regularizations, we propose a mutual contrastive learning strategy that constructs positive pairs from various network architectures and data augmentations to enhance representation learning. Extensive experiments show that our method achieves state-of-the-art performance on several cross-domain benchmarks with lightweight models, even outperforming many white-box SFDA methods.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111611"},"PeriodicalIF":7.5,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143759850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Cycle-VQA: A Cycle-Consistent Framework for Robust Medical Visual Question Answering

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-28 DOI: 10.1016/j.patcog.2025.111609

Lin Fan , Xun Gong , Cenyang Zheng , Xuli Tan , Jiao Li , Yafei Ou

引用次数: 0

FN-NET: Adaptive data augmentation network for fine-grained visual categorization

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-28 DOI: 10.1016/j.patcog.2025.111618

Shuo Ye , Qinmu Peng , Yiu-ming Cheung , Yu Wang , Ziqian Zou , Xinge You

{"title":"FN-NET: Adaptive data augmentation network for fine-grained visual categorization","authors":"Shuo Ye , Qinmu Peng , Yiu-ming Cheung , Yu Wang , Ziqian Zou , Xinge You","doi":"10.1016/j.patcog.2025.111618","DOIUrl":"10.1016/j.patcog.2025.111618","url":null,"abstract":"<div><div>Data augmentation significantly contributes to enhancing model performance, robustness, and generalization ability. However, existing methods struggle when applied directly to fine-grained targets. Particularly during perspective changes, significant details carried by local regions may be obscured or altered, making data augmentation at this point prone to severe overfitting. We argue that subclasses have common discriminative features, and these features exhibit a certain degree of complementarity. Therefore, in this paper, we propose a novel data augmentation framework for fine-grained targets called the feature expansion and noise fusion network (FN-Net). Specifically, a lightweight branch (aug-branch) is introduced in the middle layer of the convolutional neural network. Feature expansion is involved in this branch, which creates new semantic combinations from multiple instances by exchanging discriminative regions within the same subclass in the feature space. Noise fusion preserves the noise distribution of the current subclass, enhancing the model’s robustness and improving its understanding of instances in real-world environment. Additionally, to prevent potential disruptions to the original feature combinations caused by the feature expansion process, distillation loss is employed to facilitate the learning process of the aug-branch. We evaluate FN-Net on three FGVC benchmark datasets. The experimental results demonstrate that our method consistently outperforms the state-of-the-art approaches on different depths and types of network backbone structures.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111618"},"PeriodicalIF":7.5,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143768426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Automated design of neural networks with multi-scale convolutions via multi-path weight sampling

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-03-27 DOI: 10.1016/j.patcog.2025.111605

Junhao Huang , Bing Xue , Yanan Sun , Mengjie Zhang , Gary G. Yen

{"title":"Automated design of neural networks with multi-scale convolutions via multi-path weight sampling","authors":"Junhao Huang , Bing Xue , Yanan Sun , Mengjie Zhang , Gary G. Yen","doi":"10.1016/j.patcog.2025.111605","DOIUrl":"10.1016/j.patcog.2025.111605","url":null,"abstract":"<div><div>The performance of convolutional neural networks (CNNs) relies heavily on the architecture design. Recently, an increasingly prevalent trend in CNN architecture design is the utilization of ingeniously crafted building blocks, e.g., the MixConv module, for improving the model expressivity and efficiency. To leverage the feature learning capability of multi-scale convolution while further reducing its computational complexity, this paper presents a computationally efficient yet powerful module, dubbed EMixConv, by combining parameter-free concatenation-based feature reuse with multi-scale convolution. In addition, we propose a one-shot neural architecture search (NAS) method integrating the EMixConv module to automatically search for the optimal combination of the related architectural parameters. Furthermore, an efficient multi-path weight sampling mechanism is developed to enhance the robustness of weight inheritance in the supernet. We demonstrate the effectiveness of the proposed module and the NAS algorithm on three popular image classification tasks. The developed models, dubbed EMixNets, outperform most state-of-the-art architectures with fewer parameters and computations on the CIFAR datasets. On ImageNet, EMixNet is superior to a majority of compared methods and is also more compact and computationally efficient.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"165 ","pages":"Article 111605"},"PeriodicalIF":7.5,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143776791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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