Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks377 from public data
- One-step high-efficiency CRISPR/Cas9-mediated genome editing in Streptomyces291
A high-efficiency CRISPR/Cas9 genome editing plasmid pKCcas9dO for use in Streptomyces genetic manipulation, which comprises a target-specific guide RNA, a codon-optimized cas9, and two homology-directed repair templates has promise for application to genome modification in other Actinomyces species.
- Hybrid Contrastive Learning of Tri-Modal Representation for Multimodal Sentiment Analysis278
A novel framework HyCon is proposed for hybrid contrastive learning of tri-modal representation that outperforms baselines on multimodal sentiment analysis and emotion recognition and can naturally generate many training pairs for better generalization and reduce the negative effect of limited datasets.
- Modality to Modality Translation: An Adversarial Representation Learning and Graph Fusion Network for Multimodal Fusion263
A novel adversarial encoder-decoder-classifier framework to learn a modality-invariant embedding space using hierarchical graph neural network which explicitly explores unimodal, bimodal and trimodal interactions in multi-stage fusion.
- 215
- Deep multi-path convolutional neural network joint with salient region attention for facial expression recognition210
A novel model, named Deep Attentive Multi-path Convolutional Neural Network (DAM-CNN), that can automatically locate expression-related regions in an expressional image and yield a robust image representation for FER.
- Multimodal Information Bottleneck: Learning Minimal Sufficient Unimodal and Multimodal Representations184
The multi-modal information bottleneck (MIB) is introduced, aiming to learn a powerful and sufficient multimodal representation that is free of redundancy and to filter out noisy information in unimodal representations.
- Learning to Contrast the Counterfactual Samples for Robust Visual Question Answering154
This work introduces a novel self-supervised contrastive learning mechanism to learn the relationship between original samples, factual samples and counterfactual samples and evaluates the effectiveness by surpassing current state-of-the-art models on the VQA-CP dataset.
- Divide, Conquer and Combine: Hierarchical Feature Fusion Network with Local and Global Perspectives for Multimodal Affective Computing149
This work proposes a general strategy named ‘divide, conquer and combine’ for multimodal fusion, which achieves state-of-the-art performance on multi-modalities affective computing with higher efficiency.
- Communicative Message Passing for Inductive Relation Reasoning148
A Communicative Message Passing neural network for Inductive reLation rEasoning, CoMPILE, that reasons over local directed subgraph structures and has a vigorous inductive bias to process entity-independent semantic relations and can naturally handle asymmetric/anti-symmetric relations.
- Hierarchical Attention Network for Image Captioning132
A Hierarchical Attention Network (HAN) is proposed that enables attention to be calculated on pyramidal hierarchy of features synchronously and exploits several multimodal integration strategies, which can significantly improve the performance.
- Maximum Entropy Population-Based Training for Zero-Shot Human-AI Coordination116
Maximum Entropy Population-based training (MEP) is proposed to mitigate the distributional shift when paired with unencountered partners, such as humans.
- 104
- Innovative Approach for Improvement of an Antibiotic-Overproducing Industrial Strain of Streptomyces albus97
It is concluded that this aberrant protein synthesis ability in the Strr mutant, which is a result of increased stability of the 70S complex, is responsible for the remarkable salinomycin production enhancement obtained.
- Facial expression recognition with FRR‐CNN95
Different from traditional CNN, convolutional kernels of FRR-CNN is induced to be divergent by presenting a more discriminative mutual difference among feature maps of the same layer, which results in generating less redundant features and yields a more compact representation of an image.
- 94
- 92
- 89
- 88
- MissModal: Increasing Robustness to Missing Modality in Multimodal Sentiment Analysis86
A novel representation learning approach named MissModal is proposed, devoting to increasing robustness to missing modality in a classification approach, which achieves superior performance with minimal computational costs in various missing modalities scenarios (flexibility), including severely missing modality (efficiency) on two public MSA datasets.
- 86
- Orthogonal neighborhood preserving discriminant analysis for face recognition85
A new linear subspace analysis algorithm, called orthogonal neighborhood preserving discriminant analysis (ONPDA), for face recognition with consistent and promising results.
- 83
- 80
- 75
- 75
- 71
- 68
- 68
- 64
- 63
- 62
- 60
- Generative attention adversarial classification network for unsupervised domain adaptation59
This paper proposes a Generative Attention Adversarial Classification Network (GAACN) model for unsupervised domain adaptation, and presents an attention module embedded in GAN, which allows the discriminator to discriminate the transferable regions among the images of source and target domains.
- 58
- Variable lighting face recognition using discrete wavelet transform58
A new discrete wavelet transform (DWT) based illumination normalization approach for face recognition under varying lighting conditions that can preserve the illumination discontinuities when smoothing image and reduce the halo artifacts in the normalized images.
- 56
- 56
- 55
- Graph Capsule Aggregation for Unaligned Multimodal Sequences55
This paper introduces Graph Capsule Aggregation (GraphCAGE) to model unaligned multimodal sequences with graph-based neural model and Capsule Network and suggests that GraphCAGE achieves state-of-the-art performance on two benchmark datasets with representations refined by Capsule network and interpretation provided.
- 54
- 53
- 52
- 52
- An Improved Method for Semantic Image Inpainting with GANs: Progressive Inpainting51
This paper proposes an improved method named progressive inpainting, which takes a pyramid strategy from a low-resolution image to higher one, with the purpose of getting a clear completed image and reducing the reliance on the training process.
- 51
- LPF: A Language-Prior Feedback Objective Function for De-biased Visual Question Answering50
A novel Language-Prior Feedback (LPF) objective function, to re-balance the proportion of each answer's loss value in the total VQA loss, that achieves competitive performance on the bias-sensitive V QA-CP v2 benchmark.
- 48
- 48
- 47
- 47
- 46
- 46
- 45
- 41
- Bi-Directional Co-Attention Network for Image Captioning41
The proposed Bidirectional Co-Attention Network (BCAN) is able to obtain complementary information from multiple visual features via the bi-directional co-attention strategy, and integrate multimodal information via the improved multivariate residual strategy.
- Attentive matching network for few-shot learning41
This paper presents an effective framework named Attentive Matching Network (AMN) to address few-shot learning problem, and proposes a feature-level attention mechanism to help similarity function pay more emphasis on the features that better reflect the inter-class differences.
- 41
- Multiscale illumination normalization for face recognition using dual-tree complex wavelet transform in logarithm domain38
A new illumination normalization approach based on multiscale dual-tree complex wavelet transform (DT-CWT) is presented for variable lighting face recognition which can achieve satisfactory performance for recognizing face images under large illumination variations.
- 37
- Subgraph-aware Few-Shot Inductive Link Prediction via Meta-Learning37
Meta-iKG is proposed, a novel subgraph-based meta-learner for few-shot inductive relation reasoning that utilizes local subgraphs to transferSubgraph-specific information and to rapidly learn transferable patterns via meta-gradients and is evaluated on inductive benchmarks sampled from the NELL and Freebase.
- 37
- 37
- 37
- 36
- Understanding Medical Conversations with Scattered Keyword Attention and Weak Supervision from Responses36
This work approaches the medical slot filling as a multi-label classification problem with label-embedding attentive model to pay more attention to scattered medical keywords and learn the classification models by weak-supervision from responses.
- 36
- 35
- 35
- 35
- 34
- 34
- Multimodal Contrastive Learning via Uni-Modal Coding and Cross-Modal Prediction for Multimodal Sentiment Analysis34
A novel framework named MultiModal Contrastive Learning (MMCL) for multimodal representation to capture intra- and inter-modality dynamics simultaneously and designs two contrastive learning tasks, instance- and sentiment-based contrastivelearning, to promote the process of prediction and learn more interactive information related to sentiment.
- 34
- 34
- 34
- Image Inpainting Based on Patch-GANs34
The qualitative and quantitative experiments show that the proposed novel image inpainting framework has the ability to produce more realistic images compared with some existing methods, demonstrating the effectiveness and superiority of the method.
- Multimodal Graph for Unaligned Multimodal Sequence Analysis via Graph Convolution and Graph Pooling32
This work innovatively devise graph pooling algorithms to automatically explore the associations between various time slices from different modalities and learn high-level graph representation hierarchically and outperforms state-of-the-art models on three datasets under the same experimental setting.
- 32
- Constrained LSTM and Residual Attention for Image Captioning32
A model is proposed that aligns the language model to certain visual structure and also constrains it with a specific part-of-speech template and develops a residual attention mechanism to simultaneously focus on the pre-extracted visual objects and unextracted regions in an image.
- Nuclear norm based adapted occlusion dictionary learning for face recognition with occlusion and illumination changes32
A nuclear norm based error model is introduced to characterize the occlusion and corrupted region in the query image and innovatively integrate the error image with training samples to construct the dictionary, thus can both accurately reconstruct the corrupted region and non-corruption region in query images.
- 31
- 31
- Which is Making the Contribution: Modulating Unimodal and Cross-modal Dynamics for Multimodal Sentiment Analysis30
Modulation loss is designed to modulate the loss contribution based on the confidence of individual modalities in each utterance, so as to explore an optimal update solution for each unimodal network.
- 29
- Multi-Gate Attention Network for Image Captioning29
A Multi-Gate Attention (MGA) block, which expands the traditional self-attention by equipping with additional Attention Weight Gate (AWG) module and Self-Gated (SG) module, and a pre-layernorm transformer architecture to simplify the transformer architecture and make it more efficient for image feature enhancement is proposed.
- 29
- 28
- Image Captioning with Text-Based Visual Attention28
A new text-based visual attention (TBVA) model which focuses on certain salient object automatically by eliminating the irrelevant information once given previously generated text is proposed, which adopts the architecture of multimodal recurrent neural network.
- Image Captioning with Affective Guiding and Selective Attention28
A novel image captioning model with Affective Guiding and Selective Attention Mechanism named AG-SAM is proposed, which aims to bridge the affective gap between image Captioning and the emotional response elicited by the image.
- Predicting the quality of user-generated answers using co-training in community-based question answering portals28
Two semi-supervised learning methods based on co-training to predict the quality of answers by mining the relationships of non-textual features and unlabeled data in CQA are proposed.
- Illumination invariant face recognition based on dual‐tree complex wavelet transform28
By combining the obtained illumination invariant features and edge estimation information, the enhanced facial features are obtained which have more discriminating power for variable lighting face recognition.
- 27
- 27
- Deep learning‐based 3D MRI contrast‐enhanced synthesis from a 2D noncontrast T2Flair sequence27
This work demonstrates the clinical potential of synthesizing diagnostic 3D isotropic CE brain MR images from a single 2D anisotropic non-contrast sequence and achieves superior CE synthesized performance in quantitative metrics and perceptual evaluation.
- 27
- 26
- 26
- 26
- 26
- Transformer with sparse self‐attention mechanism for image captioning25
A novel transformer model with decreasing attention gates and attention fusion module is proposed, which uses attention gate to force transformer to overcome the interference of non-critical objects and capture objects information more efficiently via truncating all the attention weights that smaller than gate threshold.
- Discriminant Deep Feature Learning based on joint supervision Loss and Multi-layer Feature Fusion for heterogeneous face recognition25
A new loss function called Scatter Loss (SL) is proposed, which embeds both inter- and intra-class information for effectively training the deep model, to enhance the discriminative power of the deeply learned features.
- 25
- Image Captioning With Visual-Semantic Double Attention25
The real power of VSDA lies in the ability of not only leveraging semantic features but also eliminating the influence of irrelevant attribute words to make the semantic guidance more precise.
- Discrete ZNN models of Adams-Bashforth (AB) type solving various future problems with motion control of mobile manipulator25
Four kinds of future problems, i.e., future linear system (FLS), future division (FD), future quadratic minimization (FQM), and future equality-constrained quadratics programming (FECQP), are investigated and the efficacy and superiority of DZNN models of AB type for solving future problems are substantiated.
- Exemplar-based Cascaded Stacked Auto-Encoder Networks for robust face alignment25
A prior exemplar-based shape is utilized to handle the partial occlusion in the image so that the model can achieve robustness against local occlusions and attain higher alignment accuracy.
- 23
- Exploring influential nodes using global and local information23
An accurate and efficient algorithm for critical node mining using both global and local information (GLI) to solve the shortcoming of the existing key node identification methods that consider either local or global information.
- Aligning Image Semantics and Label Concepts for Image Multi-Label Classification23
A novel image multi-label classification framework which aims to align Image Semantics with Label Concepts (ISLC) by proposing a residual encoder to learn salient object features in the images, and exploiting the self-attention layer in aligned decoder to automatically capture the correlation between labels.
- Pseudo-margin-based universal domain adaptation23
An improved universal adaptation network (I-UAN) is proposed to perform domain alignment on the commonly-labeled samples by incorporating the pseudo-margins of target samples to estimate each source class’s probability belonging to the common label set.
- 23
- Enhanced multi‐dataset transfer learning method for unsupervised person re‐identification using co‐training strategy23
The authors' method adopts an iterative training process to improve transferred models by iterating among clustering, selection, exchange, and fine-tuning and can enhance transferred CNN models by using more source datasets and is competitive to the state-of-the-art methods.
- Cascade recurrent neural network for image caption generation23
The effectiveness of the CRNN model is verified with adopting the commonly used MSCOCO datasets, where results indicate CRNN can achieve better performance compared with the state-of-the-art image captioning methods such as Google NIC, multimodal recurrent neural network and so on.
- 22
- 22
- Moving Foreground-Aware Visual Attention and Key Volume Mining for Human Action Recognition22
A novel deep model called Moving Foreground Attention (MFA) is proposed that enhances the performance of action recognition by guiding the model to focus on the discriminative foreground targets.
- c-RNN: A Fine-Grained Language Model for Image Captioning22
A character-level RNN (c-RNN) that directly modeled on captions with characterization where descriptive sentence is composed in a flow of characters and empowered the language model to dynamically reason about word spelling as well as grammatical rules which results in expressive and elaborate sentence.
- 21
- 21
- 20
- 20
- 20
- 20
- 20
- Coverage hole detection method of wireless sensor network based on clustering algorithm20
Experimental results show that the gap Coverage performance of the proposed method is better than that of traditional method, and the gap coverage performance of this method is more suitable for the detection of coverage hole in wireless sensor network.
- TAR cloning and integrated overexpression of 6-demethylchlortetracycline biosynthetic gene cluster in Streptomyces aureofaciens20
In this study, the 6-DCT biosynthetic gene cluster (BGC) was cloned from genomic DNA of a high 6- DCT-producing strain, S. aureofaciens DM-1, using the transformation-associated recombination method.
- 19
- 19
- ICA-based neighborhood preserving analysis for face recognition19
In this paper, NPA is firstly proposed which can keep the strong discriminating power of LDA while preserving the intrinsic geometry of the in-class data samples, and IC-NPA, which combines ICA and NPA.
- 18
- 18
- Posture coordination control of two-manipulator system using projection neural network18
The posture control of the two-manipulator system is investigated and a posture coordination control scheme in the form of standard quadratic programming (QP) and a one-iteration discrete PNN (DPNN) model is proposed for convenient numerical algorithm development and digital hardware implementation.
- 18
- 18
- 17
- 17
- Gaussian Pyramid of Conditional Generative Adversarial Network for Real-World Noisy Image Denoising17
A novel pyramidal generative structural network (PGSN) is proposed for robust and efficient real-world noisy image denoising, and achieves superior performance against the state-of-the-arts.
- Facial Expression Recognition by Inter-Class Relational Learning17
This paper proposes to recognize facial expressions by extracting independent expressive representation through a novel learning procedure, called Inter-class Relational Learning (IcRL), capable of learning the mutual relations between different classes of expression and enlarging Fisher’s criterion between any two classes.
- Domain learning joint with semantic adaptation for human action recognition17
The proposed Semantic Adaptation based on the Vector of Locally Max Pooled deep learned Features (SA-VLMPF) outperforms the state-of-the-art knowledge adaptation methods.
- Influential nodes identification method based on adaptive adjustment of voting ability16
It is proved that AAV algorithm has high accuracy and effectiveness, and can be applied to real complex networks of different types and sizes.
- 16
- Learning-Based Safety-Stability-Driven Control for Safety-Critical Systems under Model Uncertainties16
A learning-based safety-stability-driven control (LBSC) algorithm is presented to guarantee the safety and tracking stability for nonlinear safety-critical systems subject to control input constraints under model uncertainties.
- 16
- Weighted Patch-based Manifold Regularization Dictionary Pair Learning model for facial expression recognition using Iterative Optimization Classification Strategy16
A novel method for solving Facial Expression Recognition (FER) task which integrates Manifold Regularization Dictionary Pair Learning and Intra-class Variation Reduced Feature model and Iterative Optimization Classification Strategy (IOCS).
- A unified hybrid transformer for joint MRI sequences super-resolution and missing data imputation15
A unified hybrid framework called SIFormer is proposed, which utilizes any available low-resolution MRI contrast configurations to complete super-resolution (SR) of poor-quality MR images and impute missing sequences simultaneously in one forward process.
- Adaptive Syncretic Attention for Constrained Image Captioning15
A novel model which simultaneously explores a better representation of images and the relationship between visual and semantic information and is on par with or even better than the state-of-the-art methods is proposed.
- TVPRNN for image caption generation15
A new time-varying parallel RNN (TVPRNN) to deal with image caption generation and shows that TVPRNN achieves better or on par with the state-of-the-art methods.
- Modified classification and regression tree for facial expression recognition with using difference expression images15
A modified classification and regression tree framework based on difference expression images, to address the facial expression recognition (FER) problem, which can simultaneously maximise intra-class purity and distance between classes, which improves the discriminating power for classification.
- Attention-Augmented Memory Network for Image Multi-Label Classification14
Experimental results on standard benchmarks, including MS-COCO 2014, PASCAL VOC 2007, and VG-500, demonstrate the effectiveness and superiority of AAMN model, which outperforms current state-of-the-art methods.
- Metric-learning-assisted domain adaptation14
A novel metric-learning-assisted domain adaptation (MLA-DA) method, which employs a novel triplet loss for helping better feature alignment and explores the relationship between the second largest probability of a target sample's prediction and its distance to the decision boundary.
- Heterogeneous Face Recognition Based on Multiple Deep Networks With Scatter Loss and Diversity Combination14
This paper designs a multiple deep networks (MDN) structure for feature extraction and proposes a joint decision strategy called diversity combination (DC) to adaptively adjust the weights of each deep network and make a joint classification decision.
- 14
- Spatiotemporal Relation Networks for Video Action Recognition14
A new end-to-end architecture called SpatioTemporal Relation Networks (STRN) to extract spatial information and temporal information simultaneously from the video with the only RGB input, which avoids the calculation of optical flow.
- Action Recognition Using Multiple Pooling Strategies of CNN Features14
Experimental results show that the three pooling strategies have complementary advantages on the challenging YouTube, UCF50 and UCF101 datasets, and the video representation is comparable to the previous state-of-the-art algorithms.
- Joint Head Attribute Classifier and Domain-Specific Refinement Networks for Face Alignment14
A two-stage refinement network is proposed for facial landmarks detection on unconstrained conditions and an adaptive sub-network training strategy applied in the DSR module can effectively solve the issue of traditional multi-view methods that an improperly selected sub- network may result in alignment failure.
- Image Captioning via Semantic Guidance Attention and Consensus Selection Strategy14
To the best knowledge, SGA-CS is the first attempt to jointly produce semantic attention guidance and select descriptive captions for image captioning tasks, achieving one of the best performance ratings among any cross-entropy training methods.
- 13
- 13
- Noise Resistible Network for Unsupervised Domain Adaptation on Person Re-Identification13
A novel unsupervised domain adaptation re-ID framework named Noise Resistible Network (NRNet), which mainly consists of two dual-stream networks that are designed with a guiding mechanism, which allows the shallow network to learn more representative feature embedding from the deep network.
- Residual attention unit for action recognition13
RAU aims to suppress the background motion by upweighting the values associated with the foreground region in the feature maps by designing a bypass to preserve the integrity of the original features by a shortcut connection between the input and output of the attention module.
- 13
- 12
- Hadamard Product Perceptron Attention for Image Captioning12
A Hadamard Product Perceptron Attention (HPPA) is proposed for image captioning, which introduces a more global feature interaction and incorporates interaction among attention heads to calculate attention results.
- m6A/ m1A /m5C/m7G-related methylation modification patterns and immune characterization in prostate cancer12
How MGRs may affect the prognosis, clinically important traits, TME, and immunotherapy responsiveness in PCa, and a very accurate nomogram is developed to improve the MRG model’s clinical applicability.
- 12
- 12
- Y-Net: Dual-branch Joint Network for Semantic Segmentation12
The traditional “U-shaped” architecture is abandoned and the proposed Y-Net is proposed, a dual-branch joint network for accurate semantic segmentation that only aggregates the high-level features with low-resolution and utilizes the global context guidance generated by the first branch to refine the second branch.
- 12
- Unsupervised Domain Adaptation via Discriminative Classes-Center Feature Learning in Adversarial Network12
A novel approach for the task of unsupervised domain adaptation via discriminative classes-center feature learning in adversarial network (C2FAN), which concentrates on learning domain-invariant representation and paying close attention to classification decision boundary simultaneously to improve the ability of transferable knowledge across different domains.
- 12
- Joint Stacked Hourglass Network and Salient Region Attention Refinement for Robust Face Alignment12
This article proposes a coarse-to-fine framework that joins a stacked hourglass network and salient region attention refinement for robust face alignment and introduces an improved Inception-ResNet unit as a basic building block.
- Facial expression recognition using intra‐class variation reduced features and manifold regularisation dictionary pair learning12
The authors generate intra-class variation reduced features (IVRF) from the difference between a query face image and its corresponding estimated image of each expression class, which can reduce negative influence from the intra- Class variations and make their model robust to intra- class variations.
- 11
- Action Recognition with Multiple Relative Descriptors of Trajectories11
This paper proposes multiple relative descriptors of trajectory to model the relative information of pairs of trajectories, and presents relative motion descriptors and relative location descriptors, which are utilized to capture the relative motion information andrelative location information respectively.
- Pseudo-3D Attention Transfer Network with Content-aware Strategy for Image Captioning11
A novel Pseudo-3D Attention Transfer network with Content-aware Strategy (P3DAT-CAS) for the image captioning task achieves very competitive performance among the state-of-the-art models.
- 11
- Application of Teaching Quality Assessment Based on Parallel Genetic Support Vector Algorithm in the Cloud Computing Teaching System11
The results of experiments show that the GA-SVM based on MapReduce is suitable for teaching quality assessment under the environment of big data.
- 11
- 11
- Direct kernel neighborhood discriminant analysis for face recognition11
A nonlinear face recognition technique based on neighborhood preserving discriminant analysis (NPDA) and a direct solution for obtaining the optimal feature vectors in feature space is presented which can preserve the most discriminative information.
- 10
- Syncretic Space Learning Network for NIR-VIS Face Recognition10
This work simultaneously synthesizes NIR and VIS images into modality-independent syncretic images and proposes a novelsyncretic space learning (SSL) model to eliminate the modal gap, and develops the Syncretic Distribution Consistency (SDC), which can enhance the intra-class compactness and learn discriminative representations.
- Curriculum Learning Meets Weakly Supervised Multimodal Correlation Learning10
This paper injects curriculum learning into weakly supervised modality correlation learning, and reaches state-of-the-art performance on MSA.
- Photorealistic Face Completion with Semantic Parsing and Face Identity-Preserving Features10
A novel face completion algorithm, called Learning and Preserving Face Completion Network (LP-FCN), which simultaneously parses face images and extracts face identity-preserving (FIP) features and can guide an identity preserving inference and ensure pixel faithfulness of completed faces.
- ADN for object detection10
A novel attentional detection network (ADN) is proposed to enrich semantic information of feature maps by adding an extra attention branch to the classic detection network and has good detection performance on different datasets without bells and whistles.
- 10
- 9
- A Feature Map is Worth a Video Frame: Rethinking Convolutional Features for Visible-Infrared Person Re-identification9
A novel Rethinking Convolutional Features (ReCF) approach for VI-ReID, which consists of two modules: Middle Feature Generation (MFG), which utilizes the feature maps in the early stage to reduce significant modality gap, and Temporal Feature Aggregation (TFA), which uses the feature Maps in the middle stage to aggregate multi-level features for enlarging the receptive field.
- 9
- 9
- Prediction of skin disease using a new cytological taxonomy based on cytology and pathology with deep residual learning method9
The results showed that the new taxonomy developed based on cytology and pathology has an overall better performance on predictive effect of skin disease than the ICD-10 on Diseases of the skin and subcutaneous tissue.
- FIN9
The proposed model can solve the problem of anchor assignment and inconsistent semantic strength between multiple detection layers mentioned above and can achieve good detection performance for objects of various sizes.
- Adaptive learning feature pyramid for object detection9
The CSFF first embeds a dilated convolution and deconvolution layer into the top-down path of the FPN to enhance the learning ability of feature fusion and an attention module is applied to suppress distraction and interference in the feature map.
- Age-Invariant Face Recognition Using Coupled Similarity Reference Coding9
By introducing non-negative constraint and coupled similarity metric, CSRC can obtain more meaningful representations and effectively enhance the performance of AIFR.
- Fine Tuning Dual Streams Deep Network with Multi-scale Pyramid Decision for Heterogeneous Face Recognition9
A novel method called fine tuning dual streams deep network (FTDSDN) with multi-scale pyramid decision (MsPD) with powerful joint decision strategy called MsPD to adaptively adjust the weight of sub structure and obtain more robust classification performance.
- DRTN: Dual Relation Transformer Network with feature erasure and contrastive learning for multi-label image classification8
A novel Dual Relation Transformer Network (DRTN) for MLIC task is presented, which surpasses current MLIC models on three challenging benchmarks, i.e., MS-COCO 2014, PASCAL VOC 2007, and NUS-WIDE datasets.
- Unsupervised Person Re-identification Using Unified Domanial Learning8
A novel unsupervised RE-ID method named unified domanial learning (UDL) is proposed, which is trained with data from a labeled source domain and an unlabeled target domain and then tested in the target domain, which resembles real-world application requirements.
- 8
- 8
- 8
- Self‐adaptive weighted synthesised local directional pattern integrating with sparse autoencoder for expression recognition based on improved multiple kernel learning strategy8
A novel method for solving facial expression recognition (FER) tasks which uses a self-adaptive weighted synthesised local directional pattern (SW-SLDP) descriptor integrating sparse autoencoder (SA) features based on improved multiple kernel learning (IMKL) strategy is presented.
- 8
- Deep Spatiotemporal Relation Learning With 3D Multi-Level Dense Fusion for Video Action Recognition8
3D Multi-Level Dense Fusion (MLDF-3D) for the deep spatiotemporal relation learning can effectively learn the correlation of two-stream features to obtain robust action representation, and achieve the state-of-the-art performance on HMDB51 and UCF101.
- Image Captioning Using Region-Based Attention Joint with Time-Varying Attention8
A novel region-based and time-varying attention network (RTAN) model for image captioning, which can determine where and when to attend to images, and can attend to only semantic information when predicting the non-visual words.
- Image Caption with Endogenous–Exogenous Attention8
This work presents a novel endogenous–exogenous attention architecture to capture both the endogenous attention, which indicates stochastic visual orienting, and the exogenous attention,Which indicates deterministic visual orientation, which achieves better performance over state-of-the-art methods.
- 8
- 7
- 7
- E3RG: Building Explicit Emotion-driven Empathetic Response Generation System with Multimodal Large Language Model7
E3RG, an Explicit Emotion-driven Empathetic Response Generation System based on multimodal LLMs which decomposes MERG task into three parts: multimodal empathy understanding, empathy memory retrieval, and multimodal response generation is proposed.
- Meta-Learn Unimodal Signals with Weak Supervision for Multimodal Sentiment Analysis7
This paper proposes a novel meta uni-label generation (MUG) framework, which leverages the available multimodal labels to learn the corresponding unimodal labels by the meta uni-label correction network (MUCN).
- 7
- End-to-end Semantic-centric Video-based Multimodal Affective Computing7
This work proposes a novel end-to-end framework named SemanticMAC to compute multimodal semantic-centric affection for human-spoken videos and presents a semantic-centric approach to unify multimodal representation learning in three ways, including gated feature interaction, multi-task pseudo label generation, and intra-/inter-sample contrastive learning.
- Multiple Temporal Pooling Mechanisms for Weakly Supervised Temporal Action Localization7
This work proposes multiple temporal pooling mechanisms (MTP) for a more sufficient information utilization, with the design of the Foreground Variance Branch, Dual Foreground Attention Branch and Hybrid Attention Fine-tuning Branch, which can leverage more effective information from different aspects and generate different CASs to guide the learning of temporal Pooling.
- Learning Joint Structure for Human Pose Estimation7
This work designs a Structural Pose Network to take full advantage of joint structure for human pose estimation under unconstrained environment and proposes a novel structure-consistency constraint, which can maintain the structural consistency between the joints and body parts via estimating whether the joints are located in their corresponding parts.
- Spatiotemporal Fusion Networks for Video Action Recognition7
This work attempts to extract efficient spatiotemporal information for video representation through a novel architecture, termed as SpatioTemporal Fusion Networks (STFN), by introducing connections between the spatial and temporal streams in two-stream networks with fusion blocks.
- Category-Level Adversaries for Semantic Domain Adaptation7
This work proposes a generative adversarial network (GAN)-based framework called category-level adversarial adaptation networks (CAA-Nets) for domain adaptation in the context of semantic segmentation and constructs an image-based generator and discriminator pair that can achieve competitive performance compared with some existing methods.
- Unsupervised Object-Level Image-to-Image Translation Using Positional Attention Bi-Flow Generative Network7
The extensive experiments validate that the proposed model is qualified to accomplish the object-level translation and obtain compelling results with other state-of-the-art approaches.
- Enhancing Sparse Decomposition Based Blade Vibration Parameter Identification7
The proposed blade parameter identification technology based on enhancing sparse decomposition is applied to different types of simulation data and rotor blade test rig to identify blades’ parameter, and the effectiveness of the algorithm has been verified.
- Visual Skeleton and Reparative Attention for Part-of-Speech image captioning system7
A novel system for image caption generation that can adapt the language models for word generation to specific syntactic structure of sentences and visual skeleton of image and achieve superior performance on COCO captioning Leaderboard is proposed.
- Joint prominent expression feature regions in auxiliary task learning network for facial expression recognition7
A novel and effective FER framework jointing prominent expression feature regions in an auxiliary task learning network (ATLN) is proposed that improves its ability to focus on regions with prominent expressions changing by sharing parameters with their auxiliary network structure.
- 6
- 6
- 6
- Excavating important nodes in complex networks based on the heat conduction model6
An algorithm for excavating important nodes based on the heat conduction model (HCM), which measures the importance of nodes by their output capacity, which is demonstrated through a comparison with eight other algorithms for excavating important nodes.
- EE-TTS: Emphatic Expressive TTS with Linguistic Information6
Ephatic Expressive TTS contains an emphasis predictor that can identify appropriate emphasis positions from text and a conditioned acoustic model to synthesize expressive speech with emphasis and linguistic information that shows strong generalization across different datasets.
- Attention-Guided Multi-Clue Mining Network for Person Re-identification6
A novel Attention-Guided Multi-Clue Mining Network (AMMN), which outperforms current competitors of relevant methods on several benchmark datasets such as Market1501, DukeMTMC-reID, CUHK03 and achieves state-of-the-art performance on Occluded datasets.
- Cascaded Structure-Learning Network with Using Adversarial Training for Robust Facial Landmark Detection6
A Cascaded Structure-Learning Network (CSLN) with using adversarial training to improve the performance of 2D facial landmark detection by taking the structure of facial landmarks into account, which achieves superior performance compared with the state-of-the-art methods.
- Curriculum Learning Meets Weakly Supervised Modality Correlation Learning6
This paper injects curriculum learning into weakly supervised multimodal correlation learning, and reaches state-of-the-art performance on MSA.
- Partial domain adaptation based on shared class oriented adversarial network6
This work proposes a shared class oriented adversarial network (SCOAN) for partial domain adaptation, and extensive experiments show that SCOAN achieves state-of-the-art results on several benchmark partialdomain adaptation datasets.
- Adaptive Interaction Modeling via Graph Operations Search6
This paper proposes to search the network structures with differentiable architecture search mechanism, which learns to construct adaptive structures for different videos to facilitate adaptive interaction modeling, and experimentally demonstrates that the designed basic graph operations in the search space are able to model different interactions in videos.
- Multivariate Attention Network for Image Captioning6
A Multivariate Attention Network (MAN) is proposed for image captioning, which contains a content attention for identifying content information of objects, a position attention for locating positions of important patches, and a minutia attention for preserving fine-grained information of target objects.
- Resetting-Label Network Based on Fast Group Loss for Person Re-Identification6
A new method, the resetting-label method, which resets the ID labels, is proposed for the Re-ID network and a fast group loss, i.e., an advanced version of variance group loss (VGL), is proposed to simplify the training process and accelerate the loss computation.
- Multi-layer Adaptive Feature Fusion for Semantic Segmentation6
A novel multi-layer adaptive feature fusion method for semantic segmentation, which is based on attention mechanism, that encourages the network to learn the importance of features from different layer according to the content of input image and the specific capability of each layer of feature, expressed in the form of weight map.
- 6
- 6
- DHC: Distributed, Hierarchical Clustering in Sensor Networks6
This paper proposes a method named Distributed, Hierarchical Clustering (DHC) for online data analysis and mining in senior networks, which clusters sensor nodes based on their current data values as well as their geographical proximity, and computes a summary for each cluster.
- 6
- Augmented DT-CWT feature based classification using Regularized Neighborhood Projection Discriminant Analysis for face recognition6
A novel face recognition method which integrates the Augmented Dual-Tree Complex Wavelet Transform representation of face images and Regularized Neighborhood Projection Discriminant Analysis (RNPDA) method, which directly obtain a set of optimal eigenvectors with a simple regression framework and can overcome the small sample size problem of NPDA.
- 6
- 5
- Dual-Active Nanoimmunomodulators for the Synergistic Enhancement of the Antitumor Efficacy of Photodynamic Immunotherapy5
CPPM enhanced the immunotherapeutic efficiency against tumors by improving the tumor immunosuppressive microenvironment, synergistically inhibiting the growth of primary and distant tumors while activating systemic antitumor immunity to eliminate lung metastases without obvious side effects.
- Multi‐level cross‐modality learning framework for text‐based person re‐identification5
An effective multi‐level cross‐modality learning (MCL) framework for language and vision person Re‐ID is proposed, designed to comprehensively map both global and partial semantic information for the visual and textual embedding at the same time.
- 5
- 5
- Boundary Adjusted Network Based on Cosine Similarity for Temporal Action Proposal Generation5
Inspired by the similarity comparison in face recognition and the similarity of action in same action segment, a module is designed to compare the similarity for visual features extracted from visual feature encoder to generate high-quality proposals.
- Mask‐guided class activation mapping network for person re‐identification5
A novel mask-guided class activation mapping (MCAM) network for person re-identi fi cation, which learns background-invariant and view-invariant features, and a new weighted channel attention (WCA) module, which replaces the global average pooling with a global depth-wise convolution layer.
- 5
- Age-Puzzle FaceNet for Cross-Age Face Recognition5
A novel model called Age-Puzzle FaceNet (APFN), based on adversarial training mechanism, that can handle age and identity attributes can provide a new insight on other robust recognition domain with respect to multi-attributes or attributes separation.
- Discriminative Face Recognition Methods with Structure and Label Information via $$l_2$$-Norm Regularization5
Three discriminative sparse representation classification methods with structure and label information based on l2,1, and locality-constrained decorrelation classification method with structure and label information can achieve better recognition results than most existing state-of-the-art sparse representation methods.
- 5
- 5
- Heterogeneous face recognition based on modality‐independent Kernel Fisher discriminant analysis joint sparse auto‐encoder5
A novel method called modality-independent Kernel discriminant analysis joint sparse auto-encoder, for solving heterogeneous face recognition problem is proposed, which does not require the data correspondences when collecting external cross-modal data and is practical for real-world cross- modal classification problem.
- 4
- Text-and-Image Learning Transformer for Cross-Modal Person Re-Identification4
A text-and-image implicit learning Transformer (TILT) is proposed to eliminate textual anisotropy and enhance the cross-modal alignment from both domains based on the bi-direction multi-modal encoders and optimize the intra-domain classification and decrease the inter-domain divergence.
- 4
- 4
- Image translation with dual‐directional generative adversarial networks4
Dual ‐ directional generative adversarial Networks are proposed, which consist of four adversarial networks, to produce images of high perceptual quality to address the problem where condition GAN produces sharp images containing many small, hallucinated objects.
- Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation4
A novel network is introduced, named multi-scale proposal regression network (MPRN), for temporal action proposal generation, which takes encoding visual features as input and predict action scores for time points, in order to group them to generate rough proposals.
- Multimodal supervised image translation4
This work proposes linearly distributed latent codes instead of conventional Gaussian vectors, which control the style of generated images and takes advantage of linear distribution, which can produce much more diverse results and outperform the state-of-the-art baselines in terms of diversity.
- Discriminative kernel sparse representation via l 2 regularisation for face recognition4
A kernel-based discriminative sparse representation method (KDSR) via l 2-norm regularisation for robust face recognition with closed-form solution that can achieve better recognition performance than many state-of-the-art methods.
- Coupled Kernel Fisher Discriminative Analysis for Low-Resolution Face Recognition4
A novel approach called coupled kernel fisher discriminative analysis (CKFDA) based on simultaneous discriminant analysis (SDA) for LR face recognition for face recognition by mapping high-resolution and low-resolution training samples into two different high-dimensional feature spaces.
- 4
- 3
- Complementary Attention Network for Weakly Supervised Temporal Action Localization3
A complementary attention network (CAN) is proposed to capture dependencies between segments and achieves excellent results on THUMOS14 and ActivityNet1.2 datasets.
- Self-Supervised Consistency Based on Joint Learning for Unsupervised Person Re-identification3
This work proposes a self-supervised consistency constraint, which uses the basic data augmentation operations without other auxiliary networks, which can improve performance of the model effectively and achieves results comparable to state-of-the-art methods.
- Attentive Part-Based Alignment Network for Vehicle Re-Identification3
A novel Attentive Part-based Alignment Network (APANet) to learn robust, diverse, and discriminative features for vehicle Re-ID, and proposes a Multi-receptive-field Attention (MA) module to adopt an efficient and cost-effective pyramid structure.
- 3
- Combining Parsing Information With Joint Structure for Human Pose Estimation3
A novel convolutional network is proposed, which consists of a pose encoder and a parsing encoder in parallel, to combine parsing information with body structure information for effectively assisting human pose estimation in unconstrained environment.
- Enhanced global attention upsample decoder based on enhanced spatial attention and feature aggregation module for semantic segmentation3
This work designs a simple but effective decoder structure based on enhanced spatial attention and feature aggregation module for semantic segmentation, which is used to recover the prediction rich in detail and semantic information.
- Selected and refined local attention module for object detection3
The authors propose a selected and refined local attention module (SRLAM) for object detection that can suppress the influence of irrelevant areas for improving detection accuracy and inspired by the classical non-local neural networks for video classification.
- 3
- Deep Captioning with Attention-Based Visual Concept Transfer Mechanism for Enriching Description3
Qualitative and quantitative results on two standard benchmarks, MSCOCO and Flickr30K show the effectiveness and practicability of the proposed A-VCTM framework.
- 3
- Regional attention generative adversarial network3
The quantitative and qualitative performance improvement over state-of-the-art methods demonstrates the validity of the proposed attention mechanism in improving the quality of generated images.
- 3
- Multimodal object description network for dense captioning3
A multimodal layer in the proposed MODN framework is adopted which can effectively extract discriminant information from both object and semantic features and can generate object description rapidly without external region proposal.
- Camera Calibration and 3D Reconstruction Using RBF Network in Stereovision System3
RBF network (RBFN) is used to provide effective methodologies for solving difficult computational problems in camera calibration and 3D reconstruction process and avoids the complicated and large calculation in conventional methods.
- Identifying influential nodes through hierarchical k-shell and extended neighborhood integration2
An algorithm named HKEN is proposed that integrates hierarchical k-shell decomposition with extended neighborhood information and incorporates both global structural features and local topological information of nodes to address the challenge of balancing accuracy and computational efficiency in existing methods.
- RCAENet: Residual Convolutional and Attention-Enhanced Stereo Matching for Real-Time Depth Estimation on Edge Devices2
This work proposes RCAENet, a high-performance stereo network designed for real-time and high-accuracy depth estimation on edge devices, and introduces the Residual Convolutional Feature Extraction (RCFE) module, which replaces conventional convolutional layers to capture more expressive features while maintaining computational efficiency.
- A Reliable Routing Algorithm Based on Path Satisfaction in the Energy Internet2
A heuristic path satisfaction-based service-aware routing algorithm (PSSRA) is designed to reduce computational complexity and demonstrates that the proposed algorithm outperforms the existing algorithms regarding the service blocking ratio and service distribution fairness index.
- 2
- Density-Aware Tree–Graph Cross-Message Passing for LiDAR Point Cloud 3D Object Detection2
This paper proposes density-aware tree–graph cross-message passing (DA-TGCMP), a unified framework that exploits the complementary strengths of both structures to enable more expressive and robust feature learning and introduces a density-aware graph construction (DAGC) strategy that adaptively models geometric relationships in regions with varying point density.
- 2
- 2
- AMVFNet: Attentive Multi-View Fusion Network for 3D Object Detection2
This article presents a novel and efficient two-stage point-pillar hybrid architecture named Attentive Multi-View Fusion Network (AMVFNet), in which features from all cylindrical view, bird-eye view, and raw point clouds are abstracted.
- 2
- 2
- Two‐way constraint network for RGB‐Infrared person re‐identification2
A novel and concise RGB-IR Re-ID network named two-way constraint network (TWCN) is proposed, which not only extracts and utilises identity-related features but also makes full use of identity-unrelated features to improve the accuracy of the experiment.
- Dual semantic interdependencies attention network for person re‐identification2
The authors proposed a dual semantic interdependencies attention module to capture semantic interependencies in both spatial dimension and channel dimension simultaneously and designed a deep supervision branch to directly guide the training of the attention modules.
- 2
- 2
- 2
- 2
- Patch-based Sparse Dictionary Representation for Face Recognition with Single Sample per Person2
A novel method patch-based sparse dictionary representation (PSDR) to tackle the problem of various variations e.g. expressions, illuminations, corruption, occlusion and disguises in FR with SSPP.
- 3D Reconstruction Approach Based on Neural Network2
In this paper, a new 3D reconstruction approach in neuro-vision system is presented and a new matching method based on Hopfield neural network (HNN) is presented, which is accurate and convincing.
- 1
- 1
- 1
- FASNet: a high-precision, two-stage stereo matching network for real-time depth perception1
FASNet is presented—a high-precision, real-time two-stage stereo matching network featuring several key innovations, including an Enhanced Residual Block with L2 normalization (ERB-L2), which replaces standard convolutions with depthwise separable convolutions and lightweight L2 normalization, enhancing feature extraction while reducing complexity.
- Semantic-Guided Multi-Feature Attention Aggregation Network for LiDAR-Based 3D Object Detection1
SMA2, a semantic-guided multi-feature attention aggregation network that adaptively integrates keypoint, voxel, and BEV features using a keypoint-guided attention mechanism, provides a more flexible and context-aware feature integration strategy.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Complementary Boundary Estimation Network for Temporal Action Proposal Generation1
Complementary Boundary Estimation Network (CBEN) is proposed, an improved approach to temporal action proposal generation based on the framework of Boundary Sensitive Network that can achieve better performance than current mainstream methods on temporalaction proposal generation.
- 1
- Weakly supervised video action localisation via two‐stream action activation network1
This Letter introduces a weakly supervised method for human action localisation, dubbed Two-stream Action Activation Network (TAAN), which fuses two-steam predictions in a simple but effective way to further improve the precision of the predictions.
- Multi-path data fusion method based on routing algorithm for wireless sensor networks1
This research presents a new method to fuse the existing data in wireless sensor networks using the current method, which simplifies the development of smart grids and increases the efficiency of the networks.
- Hierarchical extended collaborative representation based classification for single‐sample face recognition1
The authors propose a novel CRC-based FR framework that integrates hierarchical representation based model with low-rank constrained variation dictionaries and select training samples that are the nearest neighbours of test images to obtain a more discriminative training dictionary.
- Residual Gating Fusion Network for Human Action Recognition1
This paper proposes a novel convolution architecture, called Residual Gating Fusion Network (RGFN), to improve their performance by fully exploring spatiotemporal information in residual signals, and introduces Multi-Scale Convolution Fusion (MSCF) to implement spatiotsemporal fusion at multiple levels.
- 1
- 1
- 1
- Discriminative Feature Fusion with Spectral Method for Human Action Recognition1
An effective action recognition approach which differs significantly from previous interest points based approaches in that spectral information of video data is exploited and Laplacian Eigenmaps is performed on the feature space to achieve the goal of dimensionality reduction.
- 1
- 1
- 1
- 1
- 3D reconstruction approach in neurovision system1
A phase based stereo matching using Hopfield neural network approach, and a 3D object reconstruction neural network is constructed by using BP neural network, so the 3D configuration and shape can be reconstructed by this neural network.
- ST‐LoRA: SVD‐Guided Sparse Low‐Rank Adaptation With Trainable Masks for Large Language and Vision Models–
Sarse Truncated Low‐Rank Adaptation (ST‐LoRA) is introduced, a novel PEFT method that enhances LoRA through three key innovations: SVD‐based initialisation, leveraging singular value decomposition of pre‐trained weights to provide informed starting points for adaptation matrices, and Dynamic rank assignment.
- PSKNet: Lightweight kernel-aware slice network for real-time stereo depth estimation on edge devices–
- –
- –
- –
- Semantic Web-Driven Efficient Self-Attention Mechanism for High-Resolution Image Reconstruction–
The proposed semantic web-driven efficient self-attention model stacks multiple hierarchical transformer modules and expands the window size to capture feature information at different scales and introduces separation and restoration self-attention to aggregate spatial and channel information.
- Unsupervised Visible-Infrared Person ReID via Modality-Camera Balance Label Refinement–
This article proposes a hierarchical Modality-Camera Balance Label Refinement (MCBL) framework, which naturally combines Modality-Camera Balanced Label Mining, Unreliable Pseudo-Label Re-align, and Hybrid Modality-Camera Contrastive Learning into a unified framework, which balances the association information for each hierarchical domain through refining noise labels.
- –
- –
- Predicting Links in Knowledge Graphs With the Canonical Correlation Analysis and Fusing Tensor Model–
- –
- Class‐wised domain decoupling‐guided adversarial feature learning for cross‐age face recognition–
In this model, an adversarial feature learning module is designed to decompose identity features and age features, and an adversarial training strategy incorporated with an attention mechanism in the adversarial feature learning module is introduced to suppress age interference adaptively.
- –
- –
- Clinical outcomes of patients with advanced NSCLC with different EGFR exon 19 deletion subtypes treated with first-line tyrosine kinase inhibitors.–
In advanced NSCLC patients with EGFR ex19del, the efficacy of first-line TKIs therapy was affected by the presence of pleural metastasis, Ex19del subtypes and different TK is, and the mPFS of these non-classic subgroups were not statistically different with that of classic group.
- –
- Author Correction: Prediction of skin disease using a new cytological taxonomy based on cytology and pathology with deep residual learning method–
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material.
- –
- –
- –
- [Genome sequencing of Streptomyces aureofaciens DM-1 and analysis of 6-demethylchlortetracycline biosynthesis gene cluster].–
A frame-shift mutation in methyltransferase coding region was detected, which may result in the demethylation of chlortetracycline, and the complete genome sequence of S. aureofaciens DM-1 provides basic information for functional genomics studies and selection of high-yielding strains for 6-demethylchlortermine.
- OCPAN: multi‐person pose estimation with more attention on residual information–
The authors present an optimised cascaded pyramid attention network composed of two novel modules to reduce the redundant information and highlight the residual information in the bottleneck for more accurate results.
- –
- Channel and Constraint Compensation for Generative Adversarial Networks–
The channel compensation focuses on specific feature-related regions by weighting the channel of conv-layer feature maps, so specific feature modes of data distribution can be compensated and irrelevant features can also be decayed.
- –
- –
- Local Directional Amplitude Feature for Illumination Normalization with Application to Face Recognition–
This paper design an illumination robust feature by combining the directional and amplitude information as an optimal solution to the problem of face recognition under variant illumination conditions by first extracting and process the direction and amplitude Information of the pixel changes, and fuse them into a comprehensive feature.
- –
- –
- Nuclear Norm Based Superposed Collaborative Representation Classifier for Robust Face Recognition–
A superposed linear collaborative representation classifier for robust face recognition by representing the query image in terms of a superposition of the class centroid, the shared intra-class difference, and the low rank error.
- –
- –
- –
- –
- –
- –
- –
- Kernel-Based Enhanced Maximum Margin Criterion Algorithm for High-Dimensional Feature Extraction–
A new kernel discriminant analysis algorithm, called kernel-based enhanced maximum margin criterion (KEMMC), is presented for extracting features from high-dimensional data space and is extended to a nonlinear form by mapping the input space to a high- dimensional feature space that can make the mapped features linearly separable.
- –
- Application progress of heterologous expression in antibiotics research–
This review simply discusses application progress of heterologous expression in antibiotics research.
- Kernel Based Enhanced Maximum Margin Criterion for Feature Extraction–
A new kernel discriminant analysis algorithm, called Kernel-based Enhanced Maximum Margin Criterion, is presented for extracting features from high-dimensional data space by introducing a weighted matrix and taking the local property into account so that the data points of neighboring classes can be mapped far away.
- –
- Kernel Sparse Locality Preserving Canonical Correlation Analysis for Multi-modal Feature Extraction–
A kernel based sparse locality preserving canonical correlation analysis (KSLPCCA) method is presented for high dimensional feature extraction which aims to preserve the sparse reconstructive relationship of the data by minimizing a regularization-related objective function.
- Distance estimating algorithm based on hop gradient in sensor network–
An indirect distance estimation method DV-SHG (DV-hop with Smoothing Hop Gradient) is presented to improve the precision of wireless sensor network and the analysis and simulation validate that the measuring presision is improved largely at the same compute and communication cost.
- –
- Progress in research of new anti-methicillin resistant Staphylococcus aureus antibiotics–
Progress in research of new anti-MRSA antibiotics such as telavancin,ceftobiprole and ceftaroline are described.
- On the Relationship between Translation Process and Text Analysis–
The thesis argues that translation process is a process of text analysis, text analysis is aprocess of translation, and the two processes are entailed mutually.
- Network security communication and standard SSL protocol–
This paper introduces network security communication and SSL protocol in brief and puts forward improvement scheme of SSL protocol to avoid defects and summarizes SSL protocol.
- –
- –
- Modality-Interface between Semantics and Pragmatics–
This paper explores the definition and classification of modality in English language, and brings forward some assumptions about its theoretical foundation in terms of semantics and pragmatics, into which the application ofmodality is also studied.
- –
- –
- –
- APPLICATION OF NEURAL NETWORK DATA MINING TOOL TO RESEARCHING REMAINING OIL DISTRIBUTION–
The experimental results proved the validity and the feasibility of a data-mining tool based on neural network and may be used to find the breakthrough for researching remaining oil distribution.
- DESIGN OF AN OBJECT-ORIENTED NEURAL NETWORKS TOOL BASED ON MODEL–
A software engineering approach for building an artificial neural networks tool applied in petroleum domain using unified modeling language (UML) and an experiment of vacuum residue evaluation using neural networks is introduced.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.