共查询到20条相似文献，搜索用时 15 毫秒
1.
《Information processing & management》2023,60(2):103159
With the development of 3D technology and the increase in 3D models, 2D image-based 3D model retrieval tasks have drawn increased attention from scholars. Previous works align cross-domain features via adversarial domain alignment and semantic alignment. However, the extracted features of previous methods are disturbed by the residual domain-specific features, and the lack of labels for 3D models makes the semantic alignment challenging. Therefore, we propose disentangled feature learning associated with enhanced semantic alignment to address these problems. On one hand, the disentangled feature learning enables decoupling the twisted raw features into the isolated domain-invariant and domain-specific features, and the domain-specific features will be dropped while performing adversarial domain alignment and semantic alignment to acquire domain-invariant features. On the other hand, we mine the semantic consistency by compacting each 3D model sample and its nearest neighbors to further enhance semantic alignment for unlabeled 3D model domain. We give comprehensive experiments on two public datasets, and the results demonstrate the superiority of the proposed method. Especially on MI3DOR-2 dataset, our method outperforms the current state-of-the-art methods with gains of 2.88% for the strictest retrieval metric NN. 相似文献
2.
《Information processing & management》2023,60(4):103372
The unsupervised 3D model retrieval is designed to joint the information of well-labeled 2D domain and unlabeled 3D domain to learn collaborative representations. Most existing methods adopted semantic alignment, but were inevitably affected by false pseudo-label. In this paper, we design a novel Instance-Prototype Similarity Consistency Network (IPSC) to guide domain alignment with similarity consistency, which can simultaneously suppress the impact of false pseudo-label information and well reduce the domain discrepancy. IPSC contains two similarity strategies, named Single instance vs Multiple prototypes and Instance-pair vs Single prototype. The first strategy utilizes a single instance as an anchor, and measures the similarities between the anchor and multiple prototypes with the same category but from different domains. The minimization between these similarities can better align the cross-domain prototypes with Kullback–Leibler (KL) divergence than traditional Euclidean similarities. The second strategy utilizes a single prototype as an anchor, and measures the similarities between this anchor and an instance-pair with the same category but from different domains. The minimization between these similarities can conduct the instance-level alignment with KL divergence, which can better suppress the negative effect of noisy pseudo-labels. We conduct various experiments on two datasets, MI3DOR-1 (21000 2D images and 7690 3D models) and MI3DOR-2 (19694 2D images and 3982 3D models), to verify the superiority of our algorithm. 相似文献
3.
《Information processing & management》2023,60(3):103304
Person re-identification (ReID) based on heterogeneous data aims to search for the same pedestrian from different modalities. The existing unsupervised heterogeneous ReID method overly relies on pseudo labels and ignores the inter-image feature relationship. In the paper, we propose a novel Diversity Feature Constraint (DFC) method to simultaneously consider the clustering-level and instance-level feature relationship for unsupervised heterogeneous ReID. On the one hand, we employ the clustering algorithm to produce pseudo labels for heterogeneous images. Then, the clustering-level constraint is designed to optimize the model. On the other hand, considering that the clustering algorithm may generate some noise, we propose the complementary intra-modality instance-level constraint to correlate any two intra-modality images. Meanwhile, for eliminating the inter-modality discrepancy, the inter-modality instance-level constraint is developed to decrease the large inter-modality gap. We construct the potential feature relationship between heterogeneous images to constrain the feature distribution. By experiments, we prove that over-reliance on pseudo labels generates limited performance. Exploring inter-image potential relationships is an important way to solve the unsupervised problem. Extensive results demonstrate that DFC achieves superior performance that outperforms other methods by a large margin, improving 15.23% and 9.37% at rank-1 and mAP indexes compared with the clustering method on SYSU-MM01. 相似文献
4.
《Information processing & management》2020,57(5):102275
With the widespread application of 3D capture devices, diverse 3D object datasets from different domains have emerged recently. Consequently, how to obtain the 3D objects from different domains is becoming a significant and challenging task. The existing approaches mainly focus on the task of retrieval from the identical dataset, which significantly constrains their implementation in real-world applications. This paper addresses the cross-domain object retrieval in an unsupervised manner, where the labels of samples from source domain are provided while the labels of samples from target domain are unknown. We propose a joint deep feature learning and visual domain adaptation method (Deep-VDA) to solve the cross-domain 3D object retrieval problem by the end-to-end learning. Specifically, benefiting from the advantages of deep learning networks, Deep-VDA employs MVCNN for deep feature extraction and domain alignment for unsupervised domain adaptation. The framework can enable the statistical and geometric shift between domains to be minimized in an unsupervised manner, which is accomplished by preserving both common and unique characteristics of each domain. Deep-VDA can improve the robustness of object features from different domains, which is important to maintain remarkable retrieval performance. 相似文献
5.
《Information processing & management》2020,57(1):102121
Text classification or categorization is the process of automatically tagging a textual document with most relevant labels or categories. When the number of labels is restricted to one, the task becomes single-label text categorization. However, the multi-label version is challenging. For Arabic language, both tasks (especially the latter one) become more challenging in the absence of large and free Arabic rich and rational datasets. Therefore, we introduce new rich and unbiased datasets for both the single-label (SANAD) as well as the multi-label (NADiA) Arabic text categorization tasks. Both corpora are made freely available to the research community on Arabic computational linguistics. Further, we present an extensive comparison of several deep learning (DL) models for Arabic text categorization in order to evaluate the effectiveness of such models on SANAD and NADiA. A unique characteristic of our proposed work, when compared to existing ones, is that it does not require a pre-processing phase and fully based on deep learning models. Besides, we studied the impact of utilizing word2vec embedding models to improve the performance of the classification tasks. Our experimental results showed solid performance of all models on SANAD corpus with a minimum accuracy of 91.18%, achieved by convolutional-GRU, and top performance of 96.94%, achieved by attention-GRU. As for NADiA, attention-GRU achieved the highest overall accuracy of 88.68% for a maximum subsets of 10 categories on “Masrawy” dataset. 相似文献
6.
提出了一种基于光场渲染的动态3D目标重构的方法.目前基于图像建模的方法对于复杂场景难以建模,而基于图像渲染的方法因数据量大不利于动态场景的实时渲染.因此采用了模型和图像相结合的方法,从多视点视频图像中重建动态3D模型,采用光场映射算法对重建的3D模型进行光场采样,然后对样本分解压缩.实验结果表明,在保证真实感的同时,减少了数据量,并可实现动态3D目标的重构. 相似文献
7.
8.
Roya Samanipour Hamed Tahmooressi Hojatollah Rezaei Nejad Minoru Hirano Su-Royn Shin Mina Hoorfar 《Biomicrofluidics》2022,16(1)
Modern neuroscience increasingly relies on 3D models to study neural circuitry, nerve regeneration, and neural disease. Several different biofabrication approaches have been explored to create 3D neural tissue model structures. Among them, 3D bioprinting has shown to have great potential to emerge as a high-throughput/high precision biofabrication strategy that can address the growing need for 3D neural models. Here, we have reviewed the design principles for neural tissue engineering. The main challenge to adapt printing technologies for biofabrication of neural tissue models is the development of neural bioink, i.e., a biomaterial with printability and gelation properties and also suitable for neural tissue culture. This review shines light on a vast range of biomaterials as well as the fundamentals of 3D neural tissue printing. Also, advances in 3D bioprinting technologies are reviewed especially for bioprinted neural models. Finally, the techniques used to evaluate the fabricated 2D and 3D neural models are discussed and compared in terms of feasibility and functionality. 相似文献
9.
We assess the impact of R&D manpower diversity on firms’ technological performance. Relying on insights from two theoretical perspectives on team diversity (i.e. social categorization perspective and information decision-making perspective), we hypothesize that both demographic and task-related sources of diversity within firms’ R&D workforce influence technological performance. In addition, we expect that these two dimensions of diversity interact with each other. To test our hypotheses, we rely on survey data from 938 Singaporean firms, providing in-depth information on three sources of demographic diversity (i.e. gender, age, and nationality of R&D employees) and two sources of task-related diversity (i.e. educational and knowledge area background of R&D employees). Our findings point to significant interactions between different sources of R&D manpower diversity. In particular, we identify substitutive relationships between (a) educational and gender diversity, and (b) nationality and knowledge area diversity. 相似文献
10.
Daniel Carlos Guimarães Pedronette Longin Jan Latecki 《Information processing & management》2021,58(2):102443
Graph Convolutional Networks (GCNs) have been established as a fundamental approach for representation learning on graphs, based on convolution operations on non-Euclidean domain, defined by graph-structured data. GCNs and variants have achieved state-of-the-art results on classification tasks, especially in semi-supervised learning scenarios. A central challenge in semi-supervised classification consists in how to exploit the maximum of useful information encoded in the unlabeled data. In this paper, we address this issue through a novel self-training approach for improving the accuracy of GCNs on semi-supervised classification tasks. A margin score is used through a rank-based model to identify the most confident sample predictions. Such predictions are exploited as an expanded labeled set in a second-stage training step. Our model is suitable for different GCN models. Moreover, we also propose a rank aggregation of labeled sets obtained by different GCN models. The experimental evaluation considers four GCN variations and traditional benchmarks extensively used in the literature. Significant accuracy gains were achieved for all evaluated models, reaching results comparable or superior to the state-of-the-art. The best results were achieved for rank aggregation self-training on combinations of the four GCN models. 相似文献
11.
《Information processing & management》2023,60(5):103459
Semi-supervised anomaly detection methods leverage a few anomaly examples to yield drastically improved performance compared to unsupervised models. However, they still suffer from two limitations: 1) unlabeled anomalies (i.e., anomaly contamination) may mislead the learning process when all the unlabeled data are employed as inliers for model training; 2) only discrete supervision information (such as binary or ordinal data labels) is exploited, which leads to suboptimal learning of anomaly scores that essentially take on a continuous distribution. Therefore, this paper proposes a novel semi-supervised anomaly detection method, which devises contamination-resilient continuous supervisory signals. Specifically, we propose a mass interpolation method to diffuse the abnormality of labeled anomalies, thereby creating new data samples labeled with continuous abnormal degrees. Meanwhile, the contaminated area can be covered by new data samples generated via combinations of data with correct labels. A feature learning-based objective is added to serve as an optimization constraint to regularize the network and further enhance the robustness w.r.t. anomaly contamination. Extensive experiments on 11 real-world datasets show that our approach significantly outperforms state-of-the-art competitors by 20%–30% in AUC-PR and obtains more robust and superior performance in settings with different anomaly contamination levels and varying numbers of labeled anomalies. 相似文献
12.
Tingting Yang Fei Li Donghong Ji Xiaohui Liang Tian Xie Shuwan Tian Bobo Li Peitong Liang 《Information processing & management》2021,58(6):102681
Depression is a widespread and intractable problem in modern society, which may lead to suicide ideation and behavior. Analyzing depression or suicide based on the posts of social media such as Twitter or Reddit has achieved great progress in recent years. However, most work focuses on English social media and depression prediction is typically formalized as being present or absent. In this paper, we construct a human-annotated dataset for depression analysis via Chinese microblog reviews which includes 6,100 manually-annotated posts. Our dataset includes two fine-grained tasks, namely depression degree prediction and depression cause prediction. The object of the former task is to classify a Microblog post into one of 5 categories based on the depression degree, while the object of the latter one is selecting one or multiple reasons that cause the depression from 7 predefined categories. To set up a benchmark, we design a neural model for joint depression degree and cause prediction, and compare it with several widely-used neural models such as TextCNN, BiLSTM and BERT. Our model outperforms the baselines and achieves at most 65+% F1 for depression degree prediction, 70+% F1 and 90+% AUC for depression cause prediction, which shows that neural models achieve promising results, but there is still room for improvement. Our work can extend the area of social-media-based depression analyses, and our annotated data and code can also facilitate related research. 相似文献
13.
《Information processing & management》2023,60(3):103294
The paper presents new annotated corpora for performing stance detection on Spanish Twitter data, most notably Health-related tweets. The objectives of this research are threefold: (1) to develop a manually annotated benchmark corpus for emotion recognition taking into account different variants of Spanish in social posts; (2) to evaluate the efficiency of semi-supervised models for extending such corpus with unlabelled posts; and (3) to describe such short text corpora via specialised topic modelling.A corpus of 2,801 tweets about COVID-19 vaccination was annotated by three native speakers to be in favour (904), against (674) or neither (1,223) with a 0.725 Fleiss’ kappa score. Results show that the self-training method with SVM base estimator can alleviate annotation work while ensuring high model performance. The self-training model outperformed the other approaches and produced a corpus of 11,204 tweets with a macro averaged f1 score of 0.94. The combination of sentence-level deep learning embeddings and density-based clustering was applied to explore the contents of both corpora. Topic quality was measured in terms of the trustworthiness and the validation index. 相似文献
14.
通过商品市场一体化指数测度中国省域市场一体化程度,采用动态面板模型和系统GMM实证分析在密度、距离、分割的3D框架下中国省域市场一体化的影响因素。研究发现:全国市场一体化水平在波动中上升,东中西部地区呈现出"两边较低,中部偏高"的态势。在全国层面上,密度、距离因素显著推进市场一体化;财政分权阻碍了市场一体化;贸易开放对市场一体化的影响呈现出先抑制后促进的非线性特征。分地区看,部分结果显示出区域性差异,包括中西部地区产出密度阻碍了市场一体化,中部地区财政分权对市场一体化的弱化作用不显著,中西部地区贸易开放度的影响与全国整体相反等。最后,提出无差别公共政策、基础设施建设与地方干预措施等推进国内市场一体化。 相似文献
15.
Fatma Emel Kocak Bahadir Ozturk Ozben Ozden Isiklar Ozlem Genc Ali Unlu Irfan Altuntas 《Biochemia medica : ?asopis Hrvatskoga dru?tva medicinskih biokemi?ara / HDMB》2015,25(3):430-438
Introduction
Total 25-hydroxyvitamin D [25(OH)D] is the most reliable indicator of vitamin D status. In this study, we compared two automated immunoassay methods, the Abbott Architect 25-OH Vitamin D assay and the Roche Cobas Vitamin D total assay, with the liquid chromatography-tandem mass spectrometry (LC-MS/MS).Materials and methods
One hundred venous blood samples were randomly selected from routine vitamin D tests. Two of the serum aliquots were analyzed at the Abbott Architect i2000 and the Roche Cobas 6000’s module e601 in our laboratory within the same day. The other serum aliquots were analyzed at the LC-MS/MS in different laboratory. Passing-Bablok regression analysis and Bland-Altman plot were used to compare methods. Inter-rater agreement was analyzed using kappa (κ) analysis.Results
The Roche assay showed acceptable agreement with the LC-MS/MS based on Passing-Bablok analysis (intercept: -5.23 nmol/L, 95% CI: -8.73 to 0.19; slope: 0.97, 95% CI: 0.77 to 1.15). The Abbott assay showed proportional (slope: 0.77, 95% CI: 0.67 to 0.85) and constant differences (intercept: 17.08 nmol/L; 95% CI: 12.98 to 21.39). A mean bias of 15.1% was observed for the Abbott and a mean bias of -14.1% was observed for the Roche based on the Bland-Altman plots. We found strong to nearly perfect agreement in vitamin D status between the immunoassays and LC-MS/MS. (κ: 0.83 for Abbott, κ: 0.93 for Roche) using kappa analysis.Conclusion
Both immunoassays demonstrated acceptable performance, but the Roche Cobas assay demonstrated better performance than the Abbott Architect in the studied samples.Key words: 25-Hydroxyvitamin D, chromatography, immunoassay, methods, tandem mass spectrometry 相似文献16.
Sowjanya Bachali K. Dasu K. Ramalingam J. N. Naidu 《Indian journal of clinical biochemistry : IJCB》2013,28(1):74-78
Vitamin D is recognized to serve a wide range of biological functions. The presence of vitamin D receptors on different tissues explains it’s diversity of actions. Reduced levels of vitamin D is associated with insulin resistance and increased diabetes risk. The study included 50 normal healthy individuals and 49 type 2 diabetes subjects. Fasting blood glucose, total cholesterol, triglycerides, HDLc, fasting insulin, parathyroid hormone, calcium, albumin and Homeostasis model for assessment of insulin resistance (HOMAIR) were measured in all the study participants. Type 2 diabetes subjects were divided into group 1 with 25 hydroxy vitamin D (25(OH)D) ≤20 ng/ml and group 2 with 25(OH)D >20 ng/ml. By the results of this study, the mean 25(OH)D level was low (20.09 ng/ml) in type 2 diabetes compared to controls (23.89 ng/ml) and the p value was 0.02. The estimated insulin resistance by HOMAIR was more in group 1 than in group 2 of diabetes with p value of 0.037. The Pearson’s correlation-coefficient was negative for 25(OH)D and insulin in type 2 diabetes (r = ?0.294), 25(OH)D was negatively correlated with HOMAIR in total subjects. Type 2 diabetes subjects had reduced levels of vitamin D than normal individuals. The insulin resistance was more in vitamin D deficiency state. Hence vitamin D has a role in glucose metabolism, deficiency can result in insulin resistance and diabetes. 相似文献
17.
Dietmar Enko Gernot Kriegsh?user Robert Stolba Elfriede Worf Gabriele Halwachs-Baumann 《Biochemia medica : ?asopis Hrvatskoga dru?tva medicinskih biokemi?ara / HDMB》2015,25(2):203-212
Introduction
Recently several diagnostic manufacturers have launched new 25-hydroxy-vitamin D (25[OH]D) assays, which are aligned to the National Institute of Standards and Technology (NIST) Standard Reference Materials (SRM) (NIST, Gaithersburg, Maryland). The aim of this study was to compare the performance of one liquid chromatography-tandem mass spectrometry (LC-MS/MS) method, one enzyme linked immunosorbent assay (ELISA), and one recalibrated and previous version of a chemiluminescence immunoassay (CLIA).Material and methods
Serum-aliquots of 198 patient samples from routine 25(OH)D analysis were measured by the ClinMass® LC-MS/MS Complete Kit (RECIPE Chemicals + Instruments GmbH, Munich, Germany), the ORGENTEC 25(OH)D3/D2 ELISA (ORGENTEC Diagnostika GmbH, Mainz, Germany), the recalibrated Immunodiagnostic Systems (IDS)-iSYS 25(OH)DS and the previous used IDS-iSYS 25(OH)D CLIA (Immunodiagnostic Systems Ltd, Boldon, United Kingdom). Bland-Altman and Deming regression analyses were calculated for methods comparison of all tested 25(OH)D assays. The LC-MS/MS method was defined as the reference method. Within-run and between-run precision measurements were performed for all methods with three different concentration levels.Results
Compared to the LC-MS/MS method, the new IDS-iSYS 25(OH)DS and ORGENTEC 25(OH)D3/D2 assay demonstrated mean relative biases of 16.3% and 17.8%. The IDS-iSYS 25(OH)D assay showed the lowest mean bias of 1.5%. Deming regression analyses of the recalibrated IDS-iSYS 25(OH)DS and the ORGENTEC 25(OH)D3/D2 assay showed proportional differences, when compared to the reference method. All assays showed a within-run and between-run imprecision of ≤ 20% at each of the evaluated concentration levels.Conclusions
The evaluated standardized immunoassays and LC-MS/MS are useful methods for measuring 25(OH)D serum-levels in clinical laboratories.Key words: vitamin D, immunoassays, liquid chromatography-tandem mass spectrometry, reference standards, quality improvement 相似文献18.
中国海岸带土地利用遥感分类系统研究 总被引：3，自引：0，他引：3
土地利用分类是土地利用数据建立和土地利用变化研究的前提。鉴于我国海岸带土地利用多样性显著而分类系统研究相对不足的现实,本文简要论述国内外海岸带土地利用分类系统的研究进展、适用的遥感数据类型和制图精度,并从陆海耦合的角度出发回顾国内滨海湿地分类系统的研究成果。在此基础上,提出中国海岸带土地利用遥感分类系统优化方案,包含8个一级类型和24个二级类型,较全面地涵盖了全国海岸带的土地资源类型,并从海岸带区域的特征出发,强调了湿地资源的细分;应用该分类系统,基于Landsat TM影像建立解译标志,并对2010年中国海岸带区域土地利用进行遥感制图,表明该分类系统具有较强的可行性。本研究可为全国尺度海岸带土地资源遥感调查、土地利用管理与规划、海岸带综合管理等研究工作提供参考。 相似文献
19.
《Information processing & management》2023,60(4):103373
Detecting feature interactions is an important post-hoc method to explain black-box models. The literature on feature interactions mainly focus on detecting their existence and calculating their strength. Little attention has been given to the form how the features interact. In this paper, we propose a novel method to capture the form of feature interactions. First, the feature interaction sets in black-box models are detected by the high dimensional model representation-based method. Second, the pairwise separability of the detected feature interactions is determined by a novel model which is verified theoretically. Third, the set separability of the feature interactions is inferred based on pairwise separability. Fourth, the interaction form of each feature in product separable sets is explored. The proposed method not only provides detailed information about the internal structure of black-box models but also improves the performance of linear models by incorporating the appropriate feature interactions. The experimental results show that the accuracy of recognizing product separability in synthetic models is 100%. Experiments on three regression and three classification tasks demonstrate that the proposed method can capture the product separable form of feature interactions effectively and improve the prediction accuracy greatly. 相似文献
20.
Kavita K. Shalia V. K. Shah M. R. Mashru S. L. Soneji J. B. Vasvani S. Payannavar A. Walvalkar R. Mokal S. S. Mithbawkar M. Bootwalla P. Sadvekar P. K. Thakur 《Indian journal of clinical biochemistry : IJCB》2010,25(2):133-140
Matrix metalloproteinases (MMPs) play important role in the pathogenesis of coronary artery disease (CAD). 5A allele of -1612 5A/6A polymorphism of MMP-3 is associated with two fold higher activity than 6A allele. Present study was designed to analyse the association of this polymorphism with CAD in Indian population. Subjects included in the study were patients with stable angina (n=35), unstable angina (n=53), patients with recent event of myocardial infarction (MI) (MI Group-1, n=56) and patients at presentation of the acute MI (MI Group-2, n=49). Controls were healthy individuals (n=99). Genotyping of MMP-3 5A/6A polymorphism was carried out by PCR-based restriction digestion method. The genotype distribution of patient groups did not deviate from controls. Serum MMP-3 levels were significantly elevated at presentation of the acute MI by 36.8% (P=0.031) as compared to controls and more associated with 6A genotype suggesting discrepancy between in vitro transfection experiment and peripheral MMP-3 levels. 相似文献