Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks197 from public data
- 588
- 407
- Deep Sketch Hashing: Fast Free-Hand Sketch-Based Image Retrieval302
This paper introduces a novel binary coding method, named Deep Sketch Hashing (DSH), where a semi-heterogeneous deep architecture is proposed and incorporated into an end-to-end binary coding framework, and is the first hashing work specifically designed for category-level SBIR with an end to end deep architecture.
- RANet: Ranking Attention Network for Fast Video Object Segmentation241
This paper proposes a novel ranking attention module, which automatically ranks and selects these maps for fine-grained VOS performance, and develops a real-time yet very accurate Ranking Attention Network (RANet) for VOS.
- Parental harsh discipline in mainland China: Prevalence, frequency, and coexistence235
The study examined the prevalence, frequency, and coexistence of psychological aggression (PA), corporal punishment (CP), and severe physical abuse (SPA) in mainland China and highlighted the importance of studying these three types of parental harsh discipline simultaneously and intervening in harsh discipline by mothers and fathers within the same family.
- Accuracy and precision of digital volume correlation in quantifying displacements and strains in trabecular bone206
Differences as large as three-fold in accuracy and precision of the displacements and strains were found among DVC methods and among bone types, and these results demonstrate that the performance of the DVC technique can depend on trabecular structure.
- Quantization-based hashing: a general framework for scalable image and video retrieval182
Quantization-based Hashing (QBH) is a generic framework which incorporates the advantages of quantization error reduction methods into conventional property preserving hashing methods and can be applied to both unsupervised and supervised hashing methods.
- Unsupervised Deep Hashing via Binary Latent Factor Models for Large-scale Cross-modal Retrieval180
A novel unsupervised deep learning framework for multimodal data search in a self-taught manner via integrating deep learning and matrix factorization with binary latent factor models that can generate unified binary codes by solving a discrete-constrained objective function directly.
- Zero-Shot Sketch-Image Hashing175
ZSIH is the first zero- shot hashing work suitable for SBIR and cross-modal search and forms a generative hashing scheme in reconstructing semantic knowledge representations for zero-shot retrieval.
- 171
- 160
- 146
- 143
- Auto-Encoding Twin-Bottleneck Hashing133
This paper proposes an efficient and adaptive code-driven graph, which is updated by decoding in the context of an auto-encoder, and introduces into the framework twin bottlenecks that exchange crucial information collaboratively.
- Deep Asymmetric Pairwise Hashing124
This work proposes a novel Deep Asymmetric Pairwise Hashing approach (DAPH) for supervised hashing, and devise an efficient alternating algorithm to optimize the asymmetric deep hash functions and high-quality binary code jointly.
- 123
- 111
- 108
- Multiview Discrete Hashing for Scalable Multimedia Search100
This article proposes a novel unsupervised hashing method, dubbed multiview discrete hashing (MvDH), which performs matrix factorization to generate the hash codes as the latent representations shared by multiple views, during which spectral clustering is performed simultaneously.
- Performance evaluation of deep feature learning for RGB-D image/video classification94
This paper conducts extensive experiments on five popular RGB-D datasets and presents a detailed analysis about the comparison between the learned feature representations from the four deep learning models.
- 92
- The applicability of the perturbation based privacy preserving data mining for real-world data92
An individually adaptable perturbation model is proposed, which enables the individuals to choose their own privacy levels and a simple but effective and yet efficient technique to build data mining models from perturbed data is suggested.
- Parental harsh discipline and adolescent problem behavior in China: Perceived normativeness as a moderator83
The findings revealed that parental harsh discipline was associated with high levels of adolescent externalizing and internalizing behaviors, and teenagers' perceived normativeness of psychological aggression buffered the association between parental psychological aggression and adolescents' internalizing problem behaviors.
- Intergenerational transmission of harsh discipline: The moderating role of parenting stress and parent gender76
Findings revealed that both mothers' and fathers' harsh discipline were transmitted across generations and the strength of transmission varied by the severity of harsh discipline and the parent gender.
- Hierarchical prediction based on two-level Gaussian mixture model clustering for bike-sharing system75
A hierarchical prediction model that predicts the number of rents/returns to each cluster in a future period to achieve redistribution is proposed and validated on the bike-sharing system of New York City and Washington D.C.
- 74
- Unsupervised Binary Representation Learning with Deep Variational Networks67
The conditional auto-encoding variational Bayesian networks are introduced in this work to exploit the feature space structure of the training data using the latent variables, and the proposed DVB model estimates the statistics of data representations, and thus produces compact binary codes.
- A multi-label text classification method via dynamic semantic representation model and deep neural network64
A novel multi-label text classification method that combines dynamic semantic representation model and deep neural network (DSRM-DNN) is proposed that outperforms the state-of-the-art methods.
- Graph Convolutional Multi-modal Hashing for Flexible Multimedia Retrieval62
This paper proposes a Flexible Graph Convolutional Multi-modal Hashing method that adopts GCNs with linear complexity to preserve both the modality-individual and modalities-fused structural similarity for discriminative hash learning.
- Unsupervised Local Feature Hashing for Image Similarity Search62
A novel unsupervised hashing method for projecting local feature descriptors from a high-dimensional feature space to a lower-dimensional Hamming space via compact bilinear projections rather than a single large projection matrix is proposed.
- 54
- Highly-Economized Multi-view Binary Compression for Scalable Image Clustering54
This work introduces a novel approach named Highly-economized Scalable Image Clustering (HSIC) that radically surpasses conventional image clustering methods via binary compression and intuitively unify the binary representation learning and efficient binary cluster structure learning into a joint framework.
- 53
- Deep Binaries: Encoding Semantic-Rich Cues for Efficient Textual-Visual Cross Retrieval52
A novel integrated deep architecture is developed to effectively encode the detailed semantics of informative images and long descriptive sentences, named as Textual-Visual Deep Binaries (TVDB), where region-based convolutional networks with long short-term memory units are introduced to fully explore image regional details while semantic cues of sentences are modeled by a text Convolutional network.
- 51
- 50
- Set and Rebase: Determining the Semantic Graph Connectivity for Unsupervised Cross-Modal Hashing49
A novel unsupervised hashing method called Semantic-Rebased Cross-modal Hashing (SRCH) is proposed, which develops an alternating optimization routine to rebase the graph and train the hashing auto-encoders with closed-form solutions so that the overall framework is efficiently trained.
- 49
- 48
- Process of CH4-CO2 reforming over Fe/SiC catalyst under microwave irradiation47
The results indicate that the cheap Fe-based catalyst has good catalytic activity and stability under microwave irradiation and hence has a promising application.
- 46
- Fast action retrieval from videos via feature disaggregation46
A simple and efficient hashing scheme for high-dimensional video data, called Disaggregation Hashing (DH), that exploits the correlations among different feature dimensions and outperforms the state-of-the-art methods on three realistic action datasets.
- Clustering-driven unsupervised deep hashing for image retrieval44
A novel end-to-end deep framework for image retrieval, namely Clustering-driven Unsupervised Deep Hashing (CUDH), to recursively learn discriminative clusters by soft clustering model and produce binary code with high similarity responds is proposed.
- Classification by Retrieval44
A generic formulation that significantly expedites the training and deployment of image classification models, particularly under the scenarios of many image categories and high feature dimensions, and proposes a novel bit-flipping procedure which enjoys high efficacy and a local optimality guarantee.
- UniOcc: Unifying Vision-Centric 3D Occupancy Prediction with Geometric and Semantic Rendering41
This work proposes a novel Unifying Occupancy (UniOcc) prediction method, explicitly imposing spatial geometry constraint and complementing fine-grained semantic supervision through volume ray rendering, and introduces a Depth-aware Teacher Student (DTS) framework to enhance prediction accuracy using unlabeled data.
- 40
- 39
- Latent Structure Preserving Hashing38
A more generalized multi-layer LSPH (ML-LSPH) framework, in which hierarchical representations can be effectively learned by a multiplicative up-propagation algorithm, and experimental results show that ML-LSPH can achieve better performance than the single-layer LSPH and both of them outperform existing hashing techniques on large-scale data.
- Companion Paper for "MiniView Layout for Bandwidth-Efficient 360-Degree Video"33
The comparison results among the standard cube layout, the equi-angular layout (EAC), and the MiniView layout (MVL) in terms of compressed video size, visual quality of views and decoding and rendering time are reported.
- 32
- Sparse graph based self-supervised hashing for scalable image retrieval31
A sparse graph based self-supervised hashing method in which a sparse graph is constructed is proposed, which not only circumvents the requirement of a predefined number of categories as compared with pseudo label-based methods, but also significantly reduces the memory demand compared with a dense graph-based method.
- Parental corporal punishment and child anxiety in China: The moderating role of HPA-axis Activity28
The findings provide partial support for the diathesis-stress model and highlight the importance of considering how the biological vulnerability may interact with parenting to influence child adjustment.
- 28
- The Cultivating Strategies of Pre-Service Teachers' Informatization Teaching Ability Oriented to Wisdom Generation27
Through 18 weeks of teaching practice, the results show that pre-service teachers’ informatization teaching awareness, knowledge, skills and high-level abilities have generally improved.
- Embarrassingly Simple Binary Representation Learning27
This paper proposes an embarrassingly simple approach to binary representation learning, which shows that, when properly designed, even such a simple network can generate effective binary codes, by fully exploring data semantics without any held-out alternating updating steps or auxiliary models.
- 26
- 26
- Unsupervised Deep Video Hashing with Balanced Rotation26
An end-to-end hashing framework, namely Unsupervised Deep Video Hashing (UDVH), is proposed, where feature extraction, balanced code learning and hash function learning are integrated and optimized in a self-taught manner.
- 24
- Semantic-rebased cross-modal hashing for scalable unsupervised text-visual retrieval24
A novel unsupervised sparse-graph based hashing method called Semantic-Rebased Cross-modal Hashing (SRCH) is proposed, which develops an alternating optimization routine to rebase the graph and train the hashing auto-encoders with closed-form solutions so that the overall framework is efficiently trained.
- Clustering-driven Deep Adversarial Hashing for scalable unsupervised cross-modal retrieval22
A novel end-to-end deep cross-modal retrieval framework, namely Clustering-driven Deep Adversarial Hashing (CDAH), which has three main characteristics: it attempts to generate modal-invariant representations in a common space by obfuscating the modality classifier, and skillfully uses a scaled tanh function to adaptively learn the binary codes.
- Intergenerational transmission of anxiety in Chinese migrant families: The mediating role of parents’ perceptions of coparenting22
Coparenting seemed to play an important role in the intergenerational transmission of anxiety from both fathers and mothers to children in Chinese migrant families and the mediating role of both mothers' and fathers' perceptions of coparenting was evaluated.
- Deep semantic cross modal hashing with correlation alignment22
In DSCA, two deep neural networks for image and text modality separately are designed, and the semantic label information is embedded in the hash layer of the text network, which can make the learned hash matrix more stable and make the hash codes more discriminative.
- 20
- 20
- Calibration algorithm for eyetracking with unrestricted head movement20
A calibration procedure that enables researchers to track movements of the eye while allowing relatively unrestricted head and/or body movement and helps facilitate the development of cost-effective, nonproprietary eyetracking solutions.
- SplatFlow: Learning Multi-frame Optical Flow via Splatting19
An efficient MOFE framework named SplatFlow is proposed, which introduces the differentiable splatting transformation to align the previous frame’s motion feature and designs a Final-to-All embedding method to input the aligned motion feature into the current frame’s estimation, thus remodeling the existing two-frame backbones.
- 19
- 18
- Dynamically Visual Disambiguation of Keyword-based Image Search18
An adaptive multi-model framework that resolves polysemy by visual disambiguation by adapting to the dynamic changes in the search results and employing the proposed saliency-guided deep multi-instance learning network to remove outliers and learn classification models for visual disAmbiguation.
- Multi-Threshold Image Segmentation Based on Improved Grey Wolf Optimization Algorithm17
The IGWO algorithm has good global convergence and computational robustness, can effectively avoid falling into local optimum, and is especially suitable for solving high-dimensional and multi-peak complex function problems, and can be well integrated into the image segmentation process.
- 16
- 16
- CGNet: A Cascaded Generative Network for dense point cloud reconstruction from a single image16
A Cascaded Generative Network (CGNet) is proposed to reconstruct dense point clouds from a single image to preserve shape features and introduce a siamese structure to extract consistent high-level semantic from multiple images.
- Joint feature selection and graph regularization for modality-dependent cross-modal retrieval15
The key idea of JFGM is learning modality-dependent subspaces for different sub-retrieval tasks while simultaneously preserving the semantic consistency of multi-modal data.
- Application of improved grey wolf model in collaborative trajectory optimization of unmanned aerial vehicle swarm14
Aiming at the problem of unstable performance of swarm intelligence optimization algorithm under dynamic threat, deep reinforcement learning is used to optimize the model, and the model can achieve low-cost trajectory optimization, providing more reasonable technical support for unmanned aerial vehicle mission execution.
- 14
- 13
- 13
- Cross-modal dual subspace learning with adversarial network13
A novel network model termed cross-modal Dual Subspace learning with Adversarial Network (DSAN) is presented, which tries to narrow the heterogeneous gap between different modalities, and distinguish the original modality of random samples in dual subspaces.
- 11
- 11
- Semantic consistency cross-modal dictionary learning with rank constraint11
A novel Semantic Consistency cross-modal Dictionary learning algorithm with rank Constraint (SCDC) to solve the problems of heterogeneous gap reduction and effective constraints on different modalities.
- Joint Character-Level Convolutional and Generative Adversarial Networks for Text Classification10
This paper proposes a text classification framework under insufficient training sample conditions that uses the real textual features to train a generator and a discriminator so as to make the distribution of generated data consistent with that of real data.
- Intergenerational Transmission of Corporal Punishment: The Independent and Interactive Moderating Role of Children’s Negative Affectivity and Effortful Control10
Children’s NA and EC independently and interactively moderated the relations between parental childhood history of CP and their current use of CP, highlighting the importance of considering how the child factor may influence the intergenerational transmission of CP.
- 10
- Giving Text More Imagination Space for Image-text Matching9
This work proposed an imagination network to enrich the text modality based on pre-trained framework, which is helpful for image-text matching and shows the superiority and generality of the proposed framework based on two pre- trained models, CLIP and BLIP on two most frequently-used datasets MSCOCO and Flickr30K.
- Generative adversarial network based on semantic consistency for text-to-image generation9
A novel generative adversarial network based on semantic consistency is proposed to generate semantically consistent and realistic images according to text descriptions to explore the semantic consistency between text and image for an efficient cross-modal generation that combines image generation and semantic correlation.
- Robust hepatic vessels segmentation model based on noisy dataset9
A robust hepatic vessel segmentation model is built via analyzing the probability distribution relationship between noisy annotation labels and unobserved correct ones, and applied to deep neural networks (DNNs) to reduce the noise label interference.
- 9
- KSVD-Based Multiple Description Image Coding9
A new multiple description coding scheme, which is based on a sparse dictionary training method called K singular value decomposition (KSVD), where each description encodes one source subset with a small quantization stepsize, and other subsets are predictively coded with a large quantization Stepsize.
- 8
- 8
- Parental Harsh Discipline and Migrant Children’s Anxiety in China: The Moderating Role of Parental Warmth and Gender8
It was revealed that maternal but not paternal psychological aggression was found to significantly predict boys' anxiety and maternal warmth exacerbated the relations between paternal corporal punishment and girls' anxiety.
- Angular Quantization Online Hashing for Image Retrieval8
This article proposes an online hashing method called Angular Quantization Online Hashing (AQOH) to focus on learning compact binary codes with the help of cosine distance, by minimizing the quantization error between the cosine similarity calculated from the original data and the generated binary code between the existingData and the new data stream.
- 8
- 7
- 7
- 7
- 7
- Local Feature Binary Coding for Approximate Nearest Neighbor Search7
A novel supervised hashing method called Local Feature Binary Coding (LFBC) for projecting local feature descriptors from a high-dimensional feature space to a lower-dimensional Hamming space via compact bilinear projections rather than a single large projection matrix.
- 7
- 7
- L-DeeplabV3+: a lightweight semantic segmentation algorithm for complex scene perception6
The L-DeepLabV3+ model has ∼4% of the number of parameters compared to the traditional DeepLabV3+, resulting in a significant reduction in computational burden, and outperforms common semantic segmentation algorithms while using fewer model parameters and enhancing segmentation performance.
- Level set method with Retinex‐corrected saliency embedded for image segmentation6
Experiments indicate that the proposed segmentation algorithm is efficient, fast, reliable, and robust.
- Generative Reconstructive Hashing for Incomplete Video Analysis6
Generative Reconstructive Hashing is the first end-to-end framework for incomplete video analysis and outperforms the recent state-of-the-art methods by 5.44/3.22/4.82 in terms of MAPs on HMDB51/UCF101/CCV datasets, respectively.
- 5
- Video Frame Enhancement based Text Semantic Fusion for Cross-modal Text-video Retrieval5
A retrieval method which differs from traditional models that rely on complex cross-modal interactions to enhance the semantic fusion of text with video frames to improve the semantic representation of video frames during training and testing, while optimizing knowledge exploitation.
- 5
- 5
- BS3D: Building-scale 3D Reconstruction from RGB-D Images5
An easy-to-use framework for acquiring building-scale 3D reconstruction using a consumer depth camera, which enables crowd-sourcing and utilizes raw depth maps for odometry computation and loop closure refinement which results in better reconstructions.
- 5
- Weight grouping operators selection strategy for a multiobjective evolutionary algorithm based on decomposition5
A weight grouping operators selection (WGOS) strategy that picks out one operator for subsequent evolution of a multiobjective optimization problem into multiple single objective optimization subproblems for simultaneous optimization.
- 5
- Group-pair deep feature learning for multi-view 3d model retrieval5
The Group-Pair Deep Feature Learning method for multi-view 3D model retrieval is proposed, and the Margin Center Loss is given that minimizes the intra- class distance and maximize the inter-class distance.
- 5
- A discriminative training algorithm for Gaussian mixture speaker models5
A discriminative training procedure is proposed to fine-tune the parameters in the Gaussian mixture speaker model to reduce the number of misclassified vector groups.
- 4
- Unsupervised Deep Generative Hashing4
A novel unsupervised deep hashing model, named Deep Variational Binaries (DVB), with conditional auto-encoding variational Bayesian networks introduced in this work as the generative model to exploit the feature space structure of the training data using the latent variables.
- 4
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 2
- External information-augmented contrastive learning framework for fake news detection2
Extensive qualitative and quantitative experiments conducted on two real-world datasets demonstrate that EACL achieves impressive accuracy rates, significantly outperforming existing baseline methods and illustrating the effectiveness of integrating external information and contrastive learning in combating misinformation.
- 2
- Cognitive Load in Multitasking Scenarios: A Qualitative Study Overview—Analysis of Cognition and Experience2
This comprehensive review assesses qualitative research from the past decade concerning cognitive load and multitasking to underscore a complex interplay between emotional states and cognitive load, indicating that well-designed management strategies and technological supports are crucial for optimizing cognitive efficiency.
- 2
- 2
- Label projection online hashing for balanced similarity2
Lab Projection Online Hashing for Balanced Similarity overcomes the problem of similarity deviation caused by data imbalance via establishing a mapping matrix to derive a relationship between the data label vector and the binary code.
- 2
- Approximation and blind reconstruction of volumetric light field2
A blind light field reconstruction method is proposed to recover the intensity distributions and continuous volumetric data without the assistant of prior geometric information and can be further extended to other light field imaging systems if the backward image formation model can be derived.
- 2
- A New Uniform Format for 360 VR Videos2
This work introduces a new format for packing and storing 360 VR videos using two stage mappings, able to avoid internal discontinuity and generate uniform pixel distribution, while keeping the aspect ratio close to the majority standard aspect ratio of 16 : 9.
- SU‐E‐I‐87: Automated Liver Segmentation Method for CBCT Dataset by Combining Sparse Shape Composition and Probabilistic Atlas Construction2
It is demonstrated that liver structures of CBCT with artifacts can be extracted accurately for following adaptive radiation therapy.
- 2
- Multi-Scale Feature Fusion Based on Piecewise Polynomial Activation Function for Image-Text Matching1
- 1
- Multi-Scale Spatial-Temporal Feature Fusion for 3D Human Pose Estimation1
The proposed multi-scale spatial-temporal feature fusion method can improve three-dimensional human pose estimation accuracy and generate precise and smooth results while reducing computational cost and has a certain degree of generalization ability.
- Text-Guided Local Control for Regional Editing1
A novel text-based approach that enables localized control and generation of single-target images within the scope of general natural language images and enhances semantic consistency and visual appeal, offering an advancement in the field of image generation and editing.
- 1
- 1
- Learning anisotropy and asymmetry geometric features for medical image segmentation1
A new loss-function is applied to the deep learning model with dense distance regression, which can benefit the edge-based features, thus able to improve the stability of the segmentation procedure and to reduce the probability of outliers in the segmentsation results.
- 1
- Arbitrary Style Transfer with Multiple Self-Attention1
Extensive experiments show that the proposed arbitrary style transfer network can generate high-quality stylized images with better visual quality.
- 1
- Efficient DIBR method based on depth offset mapping for 3D image rendering1
An efficient DIBR method based on depth offset mapping (DOM) by calculating the normalized offset between referenced image and shifted image using only an original picture and the corresponding grayscale depth map is proposed to realize 3D display.
- Matching of 3D Model and Semantic Description via Multi-Modal Auxiliary Classifier Generative Adversarial Network With Autoencoder1
A novel network framework named Multi-modal Auxiliary Classifier Generative Adversarial Network with autoencoder (MACGAN-AE) with aims to minimize the semantic gap between 3D model and its corresponding semantic description and defines a structure-preserving loss to preserve the relationships between data after feature projection.
- 1
- 1
- Duality in Deep Reinforcement Learning--Theory1
A conceptually simple framework for deep reinforcement learning based on duality, the dual method of prioritized sampling: prioritized learning and the formula and analysis for the duality with priority are given.
- 1
- Image Formation Analysis and Light Field Information Reconstruction for Plenoptic Camera 2.01
The proposed image formation model is built based on the cognition that the plenoptic camera 2.0 is a linear forward imaging system and the method to reconstruct the light field is analyzed by solving an inverse optimization problem.
- 1
- 3D Gesture Analysis Using a Large-Scale Gesture Database1
A novel solution for real-time 3D gesture analysis using an extremely large gesture database that includes the images of various articulated hand gestures with the annotated 3D position/orientation parameters of the hand joints is introduced.
- 1
- 1
- 1
- Design and realization of distributed FTP system1
The result of application shows that the problems of multi-client requirement together is solved and transmission speed is improved, so the feasibility of distributed FTP system is testified.
- –
- –
- –
- –
- –
- Explicit semantic guided bi-incomplete multi-modal hashing with label co-occurrence and label graph constraints–
LaDiff-BIMH is proposed, an explicit semantic-guided bi-incomplete multi-modal hashing framework with label co-occurrence and label graph constraints that learns hash codes for multi-modal data with bi-incompleteness in multi-modal features and labels within a unified framework.
- –
- –
- –
- Integrating Expert Knowledge into Adversarial Model for Enhanced Text-to-Image Synthesis–
A novel that integrates Expert Knowledge into an Adversarial model (abbreviated as EKA) is proposed, utilizing the BLIP expert model to guide the discriminator and encourages the generator to produce more realistic and fine-grained images with complex semantics.
- Semantic-Consistent CLIP in Generative Adversarial Networks for Text-to-Image Synthesis–
A novel framework called Semantic-Consistent CLIP in Generative Adversarial Networks (SCC-GAN) is introduced, which presents efficient Semantic-Feature Predic- tor by leveraging the pre-trained CLIP model.
- A Unified Contrastive Framework with Multi-Granularity Fusion for Text-to-Image Generation–
A unified contrastive framework with multi-granularity fusion for text-to-image generation is proposed, and a combination of contrastive learning and the Siamese mechanism is employed to construct positive or negative pairs.
- –
- –
- –
- A multiscale circular geodesic model for image segmentation–
A new image segmentation model is proposed that combines the circular geodesic model with an adaptive cut and the multiscale image processing and can blend the benefits from both of the geodesic models and the multiscale image analysis method.
- Temporal-aware Bidirectional Attention Network for Video-Text Retrieval–
A video-text cross-modal retrieval model based on a Temporal-aware Bidirectional Attention Network (TBANet), which interacts with a similarity aggregation network to enable the retrieval model to learn high-quality text and video features is proposed.
- –
- –
- –
- Semi-Supervised Cross-Modal Retrieval Based on Discriminative Comapping–
A novel semi-supervised method for cross-modal retrieval named modal-related retrieval based on discriminative comapping (MRRDC) that makes full use of unlabelled data and demonstrates the superiority of the method compared with state-of-the-art subspace methods.
- Compound Asynchronous Exploration and Exploitation–
This work proposes an asynchronous approach to deep reinforcement learning by combining exploration and exploitation by applying a framework to off-the-shelfDeep reinforcement learning algorithms, and experimental results show that the proposed algorithm is superior in final performance and efficiency.
- Extraction of Salient Region Based on Visual Perception–
The proposed salient region extraction algorithm uses the pulse cosine transform to extract the edge information of the image, simulate the side inhibition process of the nerve cell, and obtain the image feature maps at different scales.
- Graph-Regularized NMF with Prior Knowledge for Image Inpainting–
A novel matrix factorization framework to recover the images of image inpainting by modifying the graph-regularized which allows the edge weights negative to decompose the approximation matrix.
- –
- –
- Double regularization control based on level set evolution for tablet packaging image segmentation–
Experimental studies show that the DRC method greatly improves the computational efficiency and numerical accuracy, and achieves better results for image contour segmentation compared to other level set methods.
- –
- Study on Pole in the Application of Three-Dimensional Reconstruction–
Making use of the three-dimensional reconstruction that poles features do to the two elements of the space and the quadric curve, the polar lines of the image pairs are made and the correct parameters of thespace objects are obtained.
- A Quick Method for Matching Object Subspaces Based on Visual Inspection–
A novel strategy is proposed to adopt discrete radial search paths instead of searching all points in the image, which can locate the object subspace quickly and accurately and the searching time can be reduced.
- Integrity Verification of Cloud Data–
A new program on the integrity of the data to the probabilistic methods when the number of documents inserted the pseudo-tuple becomes large and a user constructs a new type of data structure to verify its integrity in the client.
- Ontology case retrieval based on SPA and asthma diagnosis application–
Experimental results on Asthma diagnosis show that ontology based on set pair analysis case retrieval model not only reduces the amount of computation, but also improve the recall and precision.
- –
- –
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 2026-10-15. 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.