Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks391 from public data
- Deep Clustering via Joint Convolutional Autoencoder Embedding and Relative Entropy Minimization592
A new clustering model, called DEeP Embedded Regularized ClusTering (DEPICT), which efficiently maps data into a discriminative embedding subspace and precisely predicts cluster assignments is proposed, which indicates the superiority and faster running time of DEPICT in real-world clustering tasks, where no labeled data is available for hyper-parameter tuning.
- Self-Supervised Adversarial Hashing Networks for Cross-Modal Retrieval495
A self-supervised adversarial hashing (SSAH) approach, which lies among the early attempts to incorporate adversarial learning into cross-modal hashing in a self- supervised fashion, to maximize the semantic correlation and consistency of the representations between different modalities.
- 432
- Triplet-Based Deep Hashing Network for Cross-Modal Retrieval412
A triplet-based deep hashing (TDH) network for cross-modal retrieval using the triplet labels, which describe the relative relationships among three instances as supervision in order to capture more general semantic correlations between cross- modal instances.
- Multi-Task Collaborative Network for Joint Referring Expression Comprehension and Segmentation385
A novel Multi-task Collaborative Network (MCN) is proposed to achieve a joint learning of REC and RES for the first time and addresses a key challenge in this multi-task setup, i.e., the prediction conflict.
- Pairwise Relationship Guided Deep Hashing for Cross-Modal Retrieval344
This paper proposes a novel deep cross-modal hashing method to generate compact hash codes through an end-to-end deep learning architecture, which can effectively capture the intrinsic relationships between various modalities.
- Deep Spectral Clustering Using Dual Autoencoder Network328
A joint learning framework for discriminative embedding and spectral clustering is proposed, which can significantly outperform state-of-the-art clustering approaches and be more robust to noise.
- Graph Debiased Contrastive Learning with Joint Representation Clustering227
This work proposes a graph debiased contrastive learning framework, which can jointly perform representation learning and clustering, and randomly selects negative samples from the clusters which are different from the positive sample's cluster.
- A Short Survey of Recent Advances in Graph Matching223
The aim is to provide a systematic and compact framework regarding the recent development and the current state-of-the-arts in graph matching.
- 200
- 185
- Semantic Structure-based Unsupervised Deep Hashing176
This work designs a deep architecture and a pair-wise loss function to preserve the semantic structure of the semantic relationships between points in unsupervised settings, and shows that SSDH significantly outperforms current state-of-the-art methods.
- 173
- 173
- DistillHash: Unsupervised Deep Hashing by Distilling Data Pairs169
A new deep unsupervised hashing model, called DistilHash, is proposed, which can learn a distilled data set, where data pairs have confident similarity signals and the semantic similarity labels assigned by the optimal Bayesian classifier can be potentially distilled.
- Active Transfer Learning Network: A Unified Deep Joint Spectral–Spatial Feature Learning Model for Hyperspectral Image Classification168
A unified deep network, combined with active transfer learning (TL) that can be well-trained for HSIs classification using only minimally labeled training data and which significantly outperforms many state-of-the-art approaches on three popular data sets.
- Robust Reversible Watermarking via Clustering and Enhanced Pixel-Wise Masking165
This paper presents a novel pragmatic framework, wavelet-domain statistical quantity histogram shifting and clustering (WSQH-SC), which ingeniously constructs new watermark embedding and extraction procedures by histogram shift-based and histogram clustering, which are important for improving robustness and reducing run-time complexity.
- A local Tchebichef moments-based robust image watermarking164
This paper proposes a content-based watermarking scheme that combines the invariant feature extraction with watermark embedding by using Tchebichef moments to realize the robustness to common image processing operations and the blind detection.
- Adaptive-weighting discriminative regression for multi-view classification161
This work fuse the multi-view data by mapping the concatenation of original features to a discriminative low-dimensional subspace, where the features from different views are adaptively assigned with the learned optimal weights.
- Learning A Structured Optimal Bipartite Graph for Co-Clustering159
A novel co-clustering method to learn a bipartite graph with exactly k connected components, where k is the number of clusters, which approximates the original graph but maintains an explicit cluster structure.
- 158
- Shared Predictive Cross-Modal Deep Quantization157
A deep quantization approach, which is among the early attempts of leveraging deep neural networks into quantization-based cross-modal similarity search, and can preserve intramodal and intermodal similarities as much as possible and greatly reduce quantization error.
- Active multi-kernel domain adaptation for hyperspectral image classification151
A novel framework addressing HSI classification based on the domain adaptation (DA) with active learning (AL) based on utilizing the available labeled samples from source domain, and adding minimum number of the most informative samples with active queries in the target domain.
- Geometric Distortion Insensitive Image Watermarking in Affine Covariant Regions150
A geometrically invariant image watermarking based on affine covariant regions (ACRs) that provides a certain degree of robustness and is insensitive to geometric distortions as well as common image processing operations is presented.
- Graph Convolutional Network Hashing for Cross-Modal Retrieval149
A Graph Convolutional Hashing approach, which learns modality-unified binary codes via an affinity graph, and demonstrates that the proposed GCH outperforms the state-of-the-art methods.
- 147
- 142
- 136
- 134
- 130
- Coupled CycleGAN: Unsupervised Hashing Network for Cross-Modal Retrieval127
The proposed UCH seamlessly couples these two networks with generative adversarial mechanism, which can be optimized simultaneously to learn representation and hash codes, and outperforms the state-of-the-art unsupervised cross-modal hashing methods.
- 125
- 121
- 114
- 110
- Siamese Contrastive Embedding Network for Compositional Zero-Shot Learning109
This paper proposes a novel Siamese Contrastive Embedding Network (SCEN) for unseen composition recognition, and significantly outperforms the state-of-the-art approaches on three challenging benchmark datasets, including the recent proposed C-QGA dataset.
- 109
- 108
- 106
- 100
- Query-Adaptive Reciprocal Hash Tables for Nearest Neighbor Search93
Both the construction method and the query-adaptive search method are general and compatible with different types of hashing algorithms using different feature spaces and/or parameter settings and demonstrate that the proposed techniques can significantly outperform both the naive construction methods and the state-of-the-art hashing algorithms.
- 91
- Generalized Zero-Shot Learning via Disentangled Representation89
A novel method is proposed, dubbed Disentangled-VAE, which aims to disentangle category-distilling factors and category-dispersing factors from visual as well as semantic features, respectively, and outperforms the state-of-the-art approaches on four challenging benchmark datasets.
- 89
- Secure Bilevel Asynchronous Vertical Federated Learning with Backward Updating85
This work proposes a novel VFL framework integrated with new backward updating mechanism and bilevel asynchronous parallel architecture (VFB^2), under which three new algorithms, including V FB^2-SGD, -SVRG, and -SAGA, are proposed and derives the theoretical results of the convergence rates of these three algorithms under both strongly convex and nonconvex conditions.
- 84
- 80
- 79
- Adversarial Learning for Robust Deep Clustering78
Experimental results on two popular datasets show that the proposed adversarial learning method can significantly enhance the robustness and further improve the overall clustering performance.
- 77
- Lifelong Zero-Shot Learning76
This paper proposes a new ZSL setting, named as Lifelong Zero-Shot Learning (LZSL), which aims to accumulate the knowledge during the learning from multiple datasets and recognize unseen classes of all trained datasets and utilizes Variational Auto-Encoder to obtain unified semantic representations.
- Query-Adaptive Hash Code Ranking for Large-Scale Multi-View Visual Search76
This paper proposes a novel and generic approach to building multiple hash tables with multiple views and generating fine-grained ranking results at bitwise and tablewise levels, and introduces a query-adaptive bitwise weighting to alleviate the quantization loss.
- 74
- Local histogram based geometric invariant image watermarking74
Experimental results demonstrate that the proposed scheme can provide sufficient robustness against geometric attacks as well as common image processing operations.
- 72
- 72
- 72
- Attraction recommendation: Towards personalized tourism via collective intelligence66
Personalized attraction similarity (PAS) model is designed to suggest attractions through fusing heterogeneous information with weighted adaptation and simultaneously considering explicit user interaction, which is the first time that attractions are recommended by user interaction and collective intelligence in a unified framework.
- 62
- 61
- 60
- Weakly Supervised Multi-Graph Learning for Robust Image Reranking60
A novel image reranking framework is proposed by introducing a new Co-Regularized Multi- Graph Learning (Co-RMGL) framework, in which intra-graph and inter-graph constraints are integrated to simultaneously encode the similarity in a single graph and the consistency across multiple graphs.
- 59
- 59
- Context Modulated Dynamic Networks for Actor and Action Video Segmentation with Language Queries58
A context modulated dynamic convolutional operation is proposed in the proposed framework and notably outperforms state-of-the-art methods in actor and action video segmentation with language queries.
- 57
- 56
- 54
- 54
- Projection & Probability-Driven Black-Box Attack54
This paper proposes Projection & Probability-driven Black-box Attack (PPBA), a method to tackle the problem of generating adversarial examples in a black-box setting by reducing the solution space and providing better optimization.
- 52
- Content-adaptive reliable robust lossless data embedding52
This paper proposes an effective solution to RLDE by improving the histogram rotation (HR)-based embedding model and eliminates the ''salt-and-pepper'' noise in HR by the pixel adjustment mechanism, so reliable regions for embedding can be well constructed.
- Cross-Modal Learning with Adversarial Samples50
This paper proposes a novel Cross-Modal correlation Learning with Adversarial samples, namely CMLA, which for the first time presents the existence of adversarial samples in cross-modal data and provides a simple yet effective adversarial sample learning method, where inter- and intra- modality similarity regularizations across different modalities are simultaneously integrated into the learning of adversary samples.
- 49
- Hierarchical Prompt Learning for Compositional Zero-Shot Recognition48
This paper sets up three hierarchical embedding spaces that respectively model the states, the objects, and their compositions, which serve as three “experts” that can be combined in inference for more accurate predictions.
- 48
- Domain-Smoothing Network for Zero-Shot Sketch-Based Image Retrieval48
A cross-modal contrastive method is proposed to learn generalized representations to smooth the domain gap by mining relations with additional augmented samples in ZS-SBIR and a category-specific memory bank with sketch features is explored to reduce intra-class diversity in the sketch domain.
- 47
- 47
- 47
- 46
- Similarity Constraints-Based Structured Output Regression Machine: An Approach to Image Super-Resolution46
This work proposes to use structured output regression machine (SORM) to simultaneously model the inherent spatial relations between the HR and LR patches and introduces a nonlocal (NL) self-similarity prior in natural images to further enhance the SORM-based SR results.
- 45
- 44
- 43
- 43
- Deep Joint Semantic-Embedding Hashing43
A novel Deep Joint Semantic-Embedding Hashing approach that contains LabNet and ImgNet is proposed, which is conductive to the generated hash codes being more discriminative and similarity-preserving.
- Multi-task human action recognition via exploring super-category43
A novel human action recognition method based on multi-task learning framework with super-category that achieves higher accuracy with less dimensions of features over several state-of-the-art approaches is proposed.
- 41
- 39
- Group Sparse Additive Machine37
The new bound shows that GroupSAM can achieve a satisfactory learning rate with polynomial decay and Generalization error bound is derived and proved by integrating the sample error analysis with empirical covering numbers and the hypothesis error estimate with the stepping stone technique.
- 36
- 36
- 36
- Adaptive multi-bit quantization for hashing36
A multi-bit quantization method based on bit allocation to quantize each projected dimension with variable bit numbers is proposed and an incomplete coding manner by clustering to generate binary codes is proposed.
- 35
- Compressed multi-scale feature fusion network for single image super-resolution35
To solve the problems of training difficulty and computational expense consumption caused by the use of the multi-scale structure, structure sparse regularization is designed to learn a MSFF network with a sparse structure and obtain a compressed network, which greatly reduces the network parameters and accelerates the speed whilst sustaining the reconstruction quality.
- 34
- 34
- Class-Incremental Instance Segmentation via Multi-Teacher Networks32
This work proposes an incremental instance segmentation method that consists of three networks: Former Teacher Network, Current Student Network (CSN) and Current Teacher Network (CTN), which supervises FTN to preserve the previous knowledge, and CTN supervises CSN to adapt to new classes.
- New l2,1-Norm Relaxation of Multi-Way Graph Cut for Clustering32
A new relaxed multi-way graph cut clustering method, where l2,1-norm distance instead of squared distance is utilized to preserve the solution having much more clearer cluster structures, and significantly outperforms several state-of-the-art clustering approaches.
- 31
- 31
- 31
- 31
- 31
- Multi-Class Support Vector Machine via Maximizing Multi-Class Margins31
A multiclass SVM model from the perspective of maximizing margin between training points and hyperplane is proposed, and the relation between the model and other related methods is analyzed to show that the model can get better or compared results when comparing with otherrelated methods.
- Robust Sparse Coding for Mobile Image Labeling on the Cloud31
A novel mobile image labeling scheme on the cloud is presented, in which mobile images are first and efficiently transmitted to the cloud by Hamming compressed sensing, such that the heavy computation for image understanding is transferred to thecloud for quick response to the queries of the users.
- Multi-view matrix decomposition: a new scheme for exploring discriminative information31
This paper proposes a new multi-view, low-rank, and sparse matrix decomposition scheme to seamlessly integrate diverse yet complementary information stemming from multiple views and enjoys superior performance over the state-of-the-art approaches.
- Query-Adaptive Hash Code Ranking for Fast Nearest Neighbor Search31
This paper proposes a new ranking method named QRank with query-adaptive bitwise weights by exploiting both the discriminative power of each hash function and their complement for nearest neighbor search, which can achieve up to 17.11\% performance gains over state-of-the-art methods.
- Large-scale multi-task image labeling with adaptive relevance discovery and feature hashing31
A novel multi-label classification approach that seamlessly incorporates the idea of multi-task feature hashing learning that can capture the task relationships at task-level as well as feature-level.
- Adaptive Hierarchical Similarity Metric Learning With Noisy Labels30
This paper proposes an Adaptive Hierarchical Similarity Metric Learning method that considers two noise-insensitive information, i.e., class-wise divergence and sample-wise consistency, and designs an adaptive strategy to integrate this information in a unified view.
- Invisible Backdoor Attack With Dynamic Triggers Against Person Re-Identification30
A novel backdoor attack on deep ReID under a new all-to-unknown scenario, called Dynamic Triggers Invisible Backdoor Attack (DT-IBA), which instead of learning fixed triggers for the target classes from the training set, can dynamically generate new triggers for any unknown identities.
- Group Contrastive Self-Supervised Learning on Graphs30
This work proposes a group contrastive learning framework that embeds the given graph into multiple subspaces, of which each representation is prompted to encode specific characteristics of graphs, and develops principled objectives that enable us to capture the relations among both intra-space and inter-space representations in groups.
- 30
- 30
- Bilevel Distance Metric Learning for Robust Image Recognition30
This paper integrates both feature extraction and metric learning into one joint optimization framework and proposes a new bilevel distance metric learning model that characterizes the intrinsic data structure using graph regularized sparse coefficients and forces the data samples from the same class to be close to each other and pushes those from different classes far away.
- CrossFormer: Cross-Modal Representation Learning via Heterogeneous Graph Transformer29
An effective heterogeneous graph transformer for dual-stream cross-modal representation learning, named CrossFormer, is proposed, which constructs a heterogeneous graph as a bridge to achieve fine-grained intra- and inter-modal interaction on a dual-stream network.
- Exploring hybrid spatio-temporal convolutional networks for human action recognition29
This paper proposes a novel hybrid spatio-temporal convolutional network for action recognition that integrates three different type of streams into the network and can take full advantage of the spatio/temporal information of the videos.
- 28
- Adaptive graph weighting for multi-view dimensionality reduction28
This paper presents an adaptive graph weighting scheme to conduct semi-supervised multi-view dimensional reduction, and demonstrates that the proposed scheme provides a unified view to explain and understand a family of traditional schemes.
- 27
- Incremental Embedding Learning via Zero-Shot Translation27
A novel class-incremental method for embedding network, named as zero-shot translation class- increasemental method (ZSTCI), which leverages zero- shot translation to estimate and compensate the semantic gap without any exemplars in embedding networks.
- 26
- 26
- 26
- New l1-Norm Relaxations and Optimizations for Graph Clustering26
A new optimization algorithm is proposed to solve this difficult non-smooth ratio minimization problem, which involves the ratio of nonsmooth terms, and uses the l1-norm distance.
- Local Feature Based Geometric-Resistant Image Information Hiding26
Extensive experimental results and comparisons with some representative image watermarking methods confirm the excellent performance of the proposed method in robustness against various geometric distortions as well as common image processing operations.
- 25
- 25
- 25
- 25
- 24
- 24
- Faster Derivative-Free Stochastic Algorithm for Shared Memory Machines24
This paper points out the fundamental reason leading to the slow convergence rate of AsySZO, and proposes a new asynchronous stochastic zerothorder algorithm (AsySzO+), which provides a faster convergence rate O( 1 bT ) (b is the mini-batch size) by the rigorous theoretical analysis, which is a significant improvement over O(1 √ T ).
- 24
- 24
- 23
- 23
- NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens23
The design and construction of NovelQA is presented, highlighting its manual annotation, and diverse question types, and significant insights into the models’ performance are revealed, particularly emphasizing the challenges they face with multi-hop reasoning, detail-oriented questions, and extremely long input with more than 100,000 tokens.
- 23
- Vulnerability vs. Reliability: Disentangled Adversarial Examples for Cross-Modal Learning23
Novel Disentangled Adversarial examples for Cross-Modal learning is proposed, dubbed DACM, which applies the generation of adversarial perturbations to strengthen cross-modal correlations, wherein the modality-related component is acquired through gradually detaching the modalities-unrelated component.
- 22
- Environment-Invariant Curriculum Relation Learning for Fine-Grained Scene Graph Generation22
A novel Environment Invariant Curriculum Relation learning (EICR) method is proposed, which can be applied in a plug-and-play fashion to existing SGG methods to remove the imbalance of the subject-object pairs.
- Hypergraph Induced Convolutional Manifold Networks22
A hypergraph induced Convolutional Manifold Network (H-CMN) to significantly improve the representation capacity of DCNN for the complex data and achieves much better performance than the state-of-the-art.
- Multi-Level Metric Learning via Smoothed Wasserstein Distance22
A multi-level metric learning method using a smoothed Wasserstein distance to characterize the errors between any two samples, where the ground distance is considered as a Mahalanobis distance and an efficient algorithm is developed to solve the proposed new model.
- 22
- Desirable Companion for Vertical Federated Learning21
This paper reveals that zeroth-order optimization (ZOO) is a desirable companion for VFL and proposes a novel and practical VFL framework with black-box models, which is inseparably interconnected to the promising properties of ZOO.
- 21
- Deep Multi-scale Discriminative Networks for Double JPEG Compression Forensics21
A novel image tampering detection method based on deep multi-scale discriminative networks (MSD-Nets) is proposed, designed to automatically extract multiple features from the discrete cosine transform (DCT) coefficient histograms of the JPEG image.
- Deep View-Aware Metric Learning for Person Re-Identification21
A deep view-aware metric learning (DVAML) model is proposed, where image pairs with similar and dissimilar views are projected into different feature subspaces, which can discover the intrinsic relevance between image pairs from different aspects.
- 20
- 20
- 20
- Faster Stochastic Quasi-Newton Methods20
A novel faster stochastic QN method (SpiderSQN) based on the variance reduced technique of SIPDER is proposed, and it is proved that this method reaches the best known SFO complexity ofinline-formula, which also matches the existing best result.
- 20
- 19
- 19
- Towards Improved and Interpretable Deep Metric Learning via Attentive Grouping19
The proposed improved and interpretable grouping method to be integrated flexibly with any metric learning framework is based on the attention mechanism with a learnable query for each group and leads to translational invariance.
- Self-adaptive implicit contention window adjustment mechanism for QoS optimization in wireless sensor networks19
A self-adaptive implicit contention window adjustment mechanism for SMAC (Sensor MAC) is proposed based on the investigation of the medium competition behavior impacting the medium access delay and delay jitter, from the perspective of any given contending node and an optimization model of the node contention priority distribution is proposed to alleviate the aforementioned contradiction.
- Predicting Alzheimer's Disease Cognitive Assessment via Robust Low-Rank Structured Sparse Model19
This paper proposes a robust low-rank structured sparse regression method (RLSR) that simultaneously selects effective features and learns the underlying structure between cognitive scores by utilizing novel mixed structured sparsity inducing norms and low-rank approximation.
- 19
- 18
- Dynamic Reactive Spiking Graph Neural Network18
Inspired by the dynamic cognition in the brain, a Dynamic Reactive Spiking Graph Neural Network is proposed that can enhance model's expressive ability in higher biological fidelity and obtain a powerful and highly bio-fidelity model with low energy consumption.
- Improving the efficiency of identifying malignant pulmonary nodules before surgery via a combination of artificial intelligence CT image recognition and serum autoantibodies18
A novel diagnostic model of lung nodules was constructed by combining high-specific tumor markers with a high-sensitivity artificial intelligence diagnostic system that has good diagnostic performance in distinguishing malignant and benign pulmonary nodules, especially for nodules smaller than 2 cm.
- 18
- A learning-to-rank approach for image scaling factor estimation18
This paper proposes a learning-to-rank approach for automatically estimating the scaling factor based on the normalized energy density features and moment features of image scaling factor estimation, which can effectively eliminate the long-known ambiguity between upscaling and downscaling in the analysis of resampling.
- Desynchronization attacks resilient image watermarking scheme based on global restoration and local embedding18
Preliminary experiments show that the proposed feature-based image watermarking scheme can be robust against various attacks including common image processing attacks and desynchronization attacks, even complicated attacks.
- Reversible Watermarking Based on Statistical Quantity Histogram18
A novel histogram shifting based watermarking method is developed by introducing a block statistical quantity (BSQ) that reduces the diversity of grayscale histograms and guarantees the stable performance of the proposed method.
- 17
- Exploring Safety Supervision for Continual Test-time Domain Adaptation17
A soft-weighted contrastive learning module is presented to explore the highly-correlated samples and discriminate uncorrelated ones, improving the instantaneous efficiency of the model and alleviates the long-term problem of knowledge forgetting and significantly improves the accuracy of the adapted model in the late adaptation stage.
- Spatio-Temporal Catcher: A Self-Supervised Transformer for Deepfake Video Detection17
It is argued that data quality has become the main bottleneck of current research and a novel spatio-temporal generator that can synthesize various high-quality "fake" videos in large quantities at a low cost is developed, which enables the model to learn more general spatio/temporal representations in a self-supervised manner.
- 17
- A survey on multimodal-guided visual content synthesis17
This survey comprehensively summarize the improvements in multimodal-guided visual content synthesis and describes the paradigm of different modality-guided visual content synthesis, and discusses the signature methods mainly based on Generative Adversarial Networks (GANs).
- Semantic Adversarial Network with Multi-Scale Pyramid Attention for Video Classification17
A new two-stream based deep framework for video classification to discover spatial and temporal information only from RGB frames is proposed and the multi-scale pyramid attention (MPA) layer and the semantic adversarial learning (SAL) module is introduced and integrated in this framework.
- New Robust Metric Learning Model Using Maximum Correntropy Criterion17
This work proposes a new robust metric learning approach by introducing the maximum correntropy criterion to deal with real-world malicious occlusions or corruptions, and derives an efficient algorithm to solve the proposed new model and provide its convergence guarantee as well.
- Boosting Complementary Hash Tables for Fast Nearest Neighbor Search17
Extensive experiments carried out on two popular tasks including Euclidean and semantic nearest neighbor search demonstrate that the proposed boosted complementary hash-tables method enjoys the strong table complementarity and significantly outperforms the state-of-the-arts.
- Geometric Attacks Resistant Image Watermarking in Affine Covariant Regions17
A robust image watermarking in affine covariant regions is proposed, which is robust against common image processing operations, geometric attacks, and combined attacks.
- 16
- 16
- 16
- 15
- 15
- 15
- Towards Visual Feature Translation15
This paper proposes a Hybrid Auto-Encoder (HAE) to translate visual features, which learns a mapping by minimizing the translation and reconstruction errors, and an Undirected Affinity Measurement (UAM) is further designed to quantify the affinity among different types of visual features.
- Asymmetric Mutual Alignment for Unsupervised Zero-Shot Sketch-Based Image Retrieval14
A new asymmetric mutual alignment method (AMA) including a self-distillation module and a cross-modality mutual alignment module to extract the feature embeddings from unlabeled data and conducts an asymmetric strategy to update the teacher and student networks.
- 14
- Attention-guided Contrastive Hashing for Long-tailed Image Retrieval14
This paper introduces a simple yet effective model named Attention-guided Contrastive Hashing Network (ACHNet) for long-tailed hashing, where a cross attention feature enhancement module is proposed to predict the importance of features for hashing, alleviating the loss of information originated from data dimension reduction.
- 14
- Image sharpening detection based on multiresolution overshoot artifact analysis14
A novel image sharpening detection method based on multiresolution overshoot artifact analysis (MOAA) is proposed, which finds that although undergoing the same sharpening operation, the edge with large slope will present a stronger overshoot artifacts than the one with small slope.
- 14
- Histogram modification based robust image watermarking approach14
Experimental results confirm that the proposed image watermarking scheme can resist against geometric distortions as well as common image processing operations and outperforms the representative histogram-based imageWatermarking in terms of robustness.
- Reliable embedding for robust reversible watermarking14
A reliable HR method based on pixel adjustment (RHR-PA) is proposed, which modulates the range of grayscale values of pixels to remove the "salt-and-pepper" noise in HR, and to avoid the decrease of capacity in RHR-BS.
- 14
- Object and background disentanglement for unsupervised cross-domain person re-identification13
This paper proposes an unsupervised model, namely, person component decomposition and synthesis generative adversarial network (PCDS-GAN), to minimize the distribution gap among multiple person re-ID datasets.
- Robust Metric Learning on Grassmann Manifolds with Generalization Guarantees13
This work proposes a new robust formulation of metric learning that constructs a projection from higher dimensional Grassmann manifold into the one in a relative low-dimensional with more discriminative capability, where the errors between sample points are considered as an MLE (maximum likelihood estimation)-like estimator.
- Consistent Prompt Tuning for Generalized Category Discovery12
Consistent Prompt Tuning is presented to disambiguate the classes for large vision-language models (e.g., CLIP) by learning a set of “task + class” prompts for labeled and unlabeled data of both known and unknown classes.
- 12
- Capacity Assessment and Analysis of Vertiports Based on Simulation12
According to the simulation results, the vertiport operation capacity, the UAVs delay, and surface area utilization under different operation modes and topology designs are analyzed and the independent operational mode under central configuration is the optimal combination.
- 12
- 12
- Learning semantic priors for texture-realistic sketch-to-image synthesis12
This paper proposes a novel framework that explores and leverages semantic information to generate realistic textures in synthesized images for this task, and can not only synthesize images with significantly superior visual quality but is also able to achieve better results on quantitative metrics.
- Coordinate discrete optimization for efficient cross-view image retrieval12
This paper attacks the crossview hashing problem by simultaneously capturing semantic neighboring relations and maximizing the generative probability of the learned hash codes in each view, and develops a novel formulation and optimization scheme for cross-view hashing.
- 12
- 11
- 11
- 11
- 11
- 11
- Zero-shot Metric Learning11
This work tackles the zero-shot metric learning problem and proposes a novel method abbreviated as ZSML, with the purpose to learn a distance metric that measures the similarity of unseen categories (even unseen datasets).
- Performance analysis and simulation verification of S-MAC for wireless sensor networks11
This research shows two interesting results: 1) Sleeping mechanism inside S-MAC does have negative impacts on its latency and stability; fortunately, adaptive listening mechanism can optimize them to some extent; and 2) Latency and stability performances of S- MAC can be further improved through manipulating dutyCycle values.
- 10
- 10
- 10
- 10
- 10
- 10
- 10
- 9
- AceMap: Knowledge Discovery through Academic Graph9
AceMap introduces large-scale academic network visualization techniques centered on nebular graphs, providing a comprehensive view of academic networks from multiple perspectives, and proposes a unified metric based on structural entropy to quantitatively measure the knowledge content of different academic entities.
- 9
- 9
- Identifying Imaging Markers for Predicting Cognitive Assessments Using Wasserstein Distances Based Matrix Regression9
A novel robust matrix regression model with imposing Wasserstein distances on both loss function and regularization is proposed, which successfully integrate Wassersteins distance into the regression model, which can excavate the latent geometry of cognitive data.
- Assisting Attraction Classification by Harvesting Web Data9
This paper proposes a novel framework for automatically attraction classification in leveraging web-harvesting data from search engine and the photos of attractions uploaded by users, and assumes that top-k web-Harvesting images from search engines have correct labels.
- 9
- 8
- 8
- 8
- 8
- Theoretic Analysis and Extremely Easy Algorithms for Domain Adaptive Feature Learning8
This paper provides comprehensive analysis of feature learning algorithms used in conjunction with linear classifiers for domain adaptation and proposes a novel extremely easy feature learning algorithm that is extended by leveraging multiple layers, leading to a deep linear model.
- Robust curvelet-domain image watermarking based on feature matching8
Experimental results obtained using Stirmark confirm that the proposed image watermarking achieve good performance in terms of imperceptibility as well as robustness against many various distortions.
- Material-Driven Therapeutics: Functional Nanomaterial Design Paradigms Revolutionizing Osteosarcoma Treatment7
This review systematically examines nanomaterial applications in OS therapy through a materials science lens, analyzing mechanism-specific interventions and highlighting notable advancements from the past five years, establishing a framework for developing next-generation nanotherapeutic platforms to address persistent limitations in OS management.
- 7
- Knowledge-Guided Prompt Learning for Deepfake Facial Image Detection7
This work retrieves forgery-related prompts from large language models as expert knowledge to guide the optimization of learnable prompts in deepfake facial image detection, achieving significant performance improvement and boosting the application in real-world scenarios.
- Good Idea or Not, Representation of LLM Could Tell7
The findings suggest that the representations of large language models hold more potential in quantifying the value of ideas than their generative outputs, demonstrating a promising avenue for automating the idea assessment process.
- 7
- SAMControl: Controlling Pose and Object for Image Editing with Soft Attention Mask7
This article proposes a method called SAMControl (Soft Attention Mask) to adaptively control the pose and object details for image editing and utilizes a direct inversion technique to ensure the fidelity of source details within SAM.
- Entropy-Regularized Token-Level Policy Optimization for Language Agent Reinforcement7
Results show that ETPO achieves effective performance improvement on the CodeLlama-7B model and surpasses a variant PPO baseline inherited from RLHF, underlines ETPO’s potential as a robust method for refining the interactive decision-making capabilities of LLMs.
- Research on Rapid Congestion Identification Method Based on TSNE-FCM and LightGBM7
This article aims to extract and analyze the multi-scale and multi-dimensional evaluation indicators of air–ground coupled flight flow congestion, use the TSNE-FCM algorithm to classify congestion levels, and construct a real-time and fast congestion identification model using the LightGBM algorithm.
- 7
- 7
- 7
- Drosophila Gene Expression Pattern Annotations via Multi-Instance Biological Relevance Learning7
This paper computes the C2B distance by introducing class specific distance metrics and locally adaptive significance coefficients and applies the new approach to automatic gene expression pattern classification and annotation on the Drosophila melanogaster species.
- 6
- 6
- Compress to One Point: Neural Collapse for Pre-Trained Model-Based Class-Incremental Learning6
A novel Pre-Trained Model-based Class-Incremental Learning (PTM-CIL) method to explore the potential of PTM and obtain optimal class relationships, and introduces the frozen Equiangular Tight Frame classifier to construct optimal classifier structure for all seen classes, guiding the feature representation adaptation for downstream continual tasks.
- Do You Guys Want to Dance: Zero-Shot Compositional Human Dance Generation with Multiple Persons6
A novel zero-shot framework is proposed that can synthesize videos consistent with arbitrary multiple persons and background while precisely following the driving poses, and a compositional augmentation strategy to generate augmented images and utilize them to optimize a set of generalizable text embeddings is proposed.
- 6
- Label Learning Method Based on Tensor Projection6
A label learning method based on tensor projection (LLMTP), which project anchor graph into the label space through an orthogonal projection matrix to obtain cluster labels directly and introduces the tensor Schatten p-norm regularization to make the clustering label matrices of different views as consistent as possible.
- Parameter-Agnostic Deep Graph Clustering6
This article proposes the first Parameter-Agnostic Deep Graph Clustering method (PADGC), which consists of two core modules: K-guidence clustering and topological-hierarchical inference, to infer K efficiently and gain impressive clustering predictions.
- 6
- 6
- Triply stochastic gradient method for large-scale nonlinear similar unlabeled classification6
This paper proposes a scalable kernel learning algorithm for SU classification using a triply stochastic optimization framework, called TSGSU, and proves that theoretically, this method can converge to a stationary point at the rate of O(1/T) after T iterations.
- A Unified q-Memorization Framework for Asynchronous Stochastic Optimization6
An asynchronous stochastic gradient hard thresholding algorithm with q-memorization (AsySGHT-qM) for the non-convex optimization with cardinality constraint, and it is proved that the convergence rate of AsySGHT-qM before reaching the inherent error induced by gradient hard thresholding methods is geometric.
- 6
- Binarized Neural Networks for Resource-Efficient Hashing with Minimizing Quantization Loss6
This paper proposes a novel learning binary neural network framework to achieve a resource-efficient deep hashing, and provides two theories to demonstrate the necessity and effectiveness of minimizing the quantization losses for both weights and activations.
- Complementary Binary Quantization for Joint Multiple Indexing6
A complementary binary quantization (CBQ) method to jointly learning multiple hash tables that exploits the power of incomplete binary coding based on prototypes to align the original space and the Hamming space, and further utilizes the nature of multi-indexing search to jointly reduce the quantization loss based on the prototype based hash function.
- 6
- 6
- 5
- Meta-Guided Adaptive Weight Learner for Noisy Correspondence5
A novel method to achieve robust cross-modal retrieval with noisy correspondences, named the Meta Similarity Importance Assignment Network (MSIAN), that dynamically adapts the soft margin of each sample through continuously updated adaptive weights, thereby improving the robustness of the model.
- 5
- 5
- 5
- 5
- 5
- EvolveKG: a general framework to learn evolving knowledge graphs5
A general framework that enables algorithms in the static knowledge graphs to learn the evolving ones and quantifies the influence of a historical fact on a current one, called the effectiveness of the fact, and makes knowledge prediction by leveraging all the cross-time knowledge interaction.
- 5
- MSR: Making Self-supervised learning Robust to Aggressive Augmentations5
This work proposes a new SSL paradigm, which counteracts the impact of semantic shift by balancing the role of weak and aggressively augmented pairs, and can better embrace the aggressive augmentations and neutralize the semantic shift problem.
- Orthogonality-Promoting Dictionary Learning via Bayesian Inference5
A novel robust dictionary learning framework via Bayesian inference that employs the orthogonality-promoting regularization to mitigate correlations among dictionary atoms and can learn the dictionary with an accuracy better than existing methods, especially when the number of training signals is limited.
- Joint Coupled-Hashing Representation for Cross-Modal Retrieval5
A novel cross-modal hashing method, which embeds heterogeneous features into their respective Hamming spaces and the correlations between different modalities are maximized by joint coupled-hashing representation.
- Coupled Dictionary Learning with Common Label Alignment for Cross-Modal Retrieval5
This work proposes a novel cross-modal retrieval method based on coupled dictionary learning with common label alignment that outperforms several state-of-the-art methods.
- 5
- 5
- 4
- Navigating Continual Test-time Adaptation with Symbiosis Knowledge4
A Dual-stream Network that independently optimizes different parameters in each stream to capture symbiotic knowledge from continual domains, thereby ensuring generalization while enhancing instantaneous discrimination and to prevent catastrophic forgetting is introduced.
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- How Unlabeled Web Videos Help Complex Event Detection?4
A new robust dictionary learning framework for complex event detection, able to handle both labeled and easy-to-get unlabeled web videos by sharing the same dictionary is proposed, which shows strong robustness against the substantial noisy and outlier videos from open source.
- 4
- 4
- Discriminative Latent Feature Space Learning for Cross-Modal Retrieval4
A general model which jointly learns a discriminative latent feature space for effective cross-modal retrieval is proposed, and the experimental results confirmed that the cross- modal method outperforms several competing methods on two public datasets.
- Image super-resolution using multi-layer support vector regression4
A novel image SR approach is presented, where a multi-layer SVR model is adopted to describe the relationship between the low resolution image patches and the corresponding high resolution (HR) ones, and pixel-wise classification is introduced to divide pixels into different classes.
- 4
- Image super resolution using Gaussian Process Regression with patch clustering4
This work proposes a novel image super resolution approach based on Gaussian Process Regression, and utilizes K-means clustering algorithm to group all the training pixel-patch samples according to the local geometric structure of LR patches.
- Content-based Image Retrieval Using Combination Features4
Improved thoughts are proposed and they are useful to overcome many shortcomings of the primal method and will help CBIR to be able to cope with the demands of the rapidly changing environment.
- 3
- 3
- A Bridge from Audio to Video: Phoneme-Viseme Alignment Allows Every Face to Speak Multiple Languages3
Multilingual Experts (MuEx) is proposed, a novel framework featuring a Phoneme-Guided Mixture-of-Experts (PG-MoE) architecture that employs phonemes and visemes as universal intermediaries to bridge the gap between audio and visual modalities, enabling lifelike multilingual TFS.
- 3
- AStF: Motion Style Tranfer via Adaptive Statistics Fusor3
A novel Adaptive Statistics Fusor which consists of Style Disentanglement Module (SDM) and High-Order Multi-Statistics Attention (HOS-Attn) and shows proficiency superiority in motion style transfers over state-of-the-arts.
- 3
- Energy vs. Noise: Towards Robust Temporal Action Localization in Open-World3
The Energy-Driven Meta Purifier (EDMP) method is proposed, which utilizes a meta-learning training paradigm to avoid dependence on extensive and precise manual annotations, and uses energy modeling to distinguish between different actions and backgrounds from the perspective of energy differences, thereby improving the model's robustness to category noise.
- AI-Generated Image Quality Assessment Based on Task-Specific Prompt and Multi-Granularity Similarity3
- Self-Supervised Graph Embedding Clustering3
A novel self-supervised manifold clustering framework is proposed that reformulates the two models into a unified framework, eliminating the need for additional hyperparameters while achieving dimensionality reduction clustering.
- 3
- 3
- FMGNN: Fused Manifold Graph Neural Network3
The Fused Manifold Graph Neural Network (FMGNN), a novel GNN architecture that embeds graphs into different Riemannian manifolds with interaction and alignment among these manifolds during training and fuses the vertex embeddings through the distances on different manifolds between vertices and selected landmarks, geometric coresets, is proposed.
- Covidia: COVID-19 Interdisciplinary Academic Knowledge Graph3
This work designs frameworks based on contrastive learning for disciplinary classification, and proposes a new academic knowledge graph scheme for entity extraction, relation classification and ontology management in accordance with interdisciplinary researches.
- 3
- Dictionary Learning in Optimal Metric Space3
This work proposes a new unified unsupervised model which naturally integrates metric learning to enhance dictionary learning model with fully utilizing the side information, and updates metric space and dictionary adaptively and alternatively, which ensures learning optimal metricspace and dictionary simultaneously.
- Fast Road Vanishing Point Detection Based on Modified Adaptive Soft Voting3
A framework for vanishing point detection based on the Modified Adaptive Soft Voting (MASV) scheme is proposed, which outperforms by 4.64% on an average than the complete-map based gLoG, which is the state-of-the-art method.
- Image Super-resolution Based on Sparse Representation With Joint Constraints3
A novel image SR algorithm based on sparse representation with joint constraints that outperforms the state-of-the-arts and gradually magnify the LR input in the current frame to the desired size is proposed.
- Query-dependent visual dictionary adaptation for image reranking3
A query-dependent image reranking approach by leveraging the higher level attribute detection among the top returned images to adapt the dictionary built over the visual features to a query-specific fashion is proposed.
- 3
- Activation-aware Probe-Query: Effective Key-Value Retrieval for Long-Context LLMs Inference2
ActQKV is proposed, a training-free, training-free, Activation-aware approach that dynamically determines probe-Query and leverages it to retrieve the relevant KV pairs for inference, and designs a dynamic KV cut-off mechanism guided by information density across layers at the decoding stage.
- 2
- Tackling Long-Tailed Data Challenges in Spiking Neural Networks via Heterogeneous Knowledge Distillation2
This paper proposes a long-tail learning framework based on spiking neural networks, named LT-SpikingFormer, to alleviate the distribution bias between head and tail classes, and adopts a widely trained Convolutional Neural Network to construct a heterogeneous knowledge distillation paradigm.
- Outstanding Orthodontist: No More Artifactual Teeth in Talking Face2
OrthoNet is proposed, a plug-and-play framework designed to eliminate unrealistic teeth effects in audio-driven TFS and acts like an orthodontist for existing Audio2Video methods, ensuring that teeth maintain natural rigidity and temporal consistency even under varying degrees of teeth occlusion.
- 2
- Accuracy of next-generation sequencing for molecular profiling of small specimen of lung cancer: a prospective pilot study of side-by-side comparison2
Pulmonary CNB specimens were suitable for NGS given the satisfactory accuracy when compared to corresponding surgical specimens, and should be deemed reliable to provide instructive information for the treatment of advanced lung cancers.
- 2
- 2
- 2
- Setting strategy of delay-optimization-oriented SMAC contention window size2
This paper proposes mathematical models that allow the analysis of data packets forwarding delay within one SMAC virtual cluster, and demonstrates that the proposed setting strategy makes contributions to the improvement in the existing SMAC extensions when they are integrated with each other, in terms of the end-to-end delay, throughput and energy consumption.
- Optimizing Bag Features for Multiple-Instance Retrieval2
An embedded scheme is devised as encoding each bag into a single bag feature vector by exploiting a similarity-based transformation and a principled approach for optimizing bag features specific to similarity retrieval through leveraging pairwise label information at the bag level is developed.
- Encryption for Image Watermarking Based on Multiple Cellular Automata2
Experimental results illustrate that the proposed encryption method is highly secure in terms of key space, histogram distribution, correlation, information entropy and gray scale change, and that it effectively solves the security problems of RW-HS.
- AN IMAGE QUALITY METRIC BASED ON BIOLOGICALLY INSPIRED FEATURE MODEL2
Experimental results confirm that the proposed IQA metric improves the performance over PSNR and SSIM under various distortion types in terms of different evaluation criteria.
- Color image quality assessment with biologically inspired feature and machine learning2
A biologically inspired model is adopted to mimic the visual cortex and represent a color image based on BIFs which unifies color units, intensity units and C1 units and the method classifies the quality according to the quality scale recommended by the ITU.
- The Modelling of Real Hair Based on B-Spline2
The result of simulation demonstrates the method is simple in the controlling of hair, and can create various hair style.
- CAMS: Towards Compositional Zero-Shot Learning via Gated Cross-Attention and Multi-Space Disentanglement1
CAMS is proposed, which aims to extract semantic features from visual features and perform semantic disentanglement in multidimensional spaces, thereby improving generalization over unseen attribute-object compositions and achieves state-of-the-art performance in both closed-world and open-world settings.
- Discriminative Anchor Learning for Hyperspectral Image Clustering1
By sharing the coefficient matrix among anchors and that between samples and anchors, the proposed method explicitly takes into account the distribution of samples and anchors, enabling the learned anchors to better describe the samples.
- High-Discriminative Attribute Feature Learning for Generalized Zero-Shot Learning1
High-Discriminative Attribute Feature Learning for Generalized Zero-Shot Learning optimizes visual features by learning attribute features to obtain discriminative visual embeddings and introduces a Transformer-based attribute discrimination encoder to enhance the discriminative capability among attributes.
- ACFun: Abstract-Concrete Fusion Facial Stylization1
A novel method called ACFun is proposed, which uses only one style image and one facial image for facial stylization, and outperforms others in facial stylization, producing highly artistic and visually pleasing results.
- Interpretable Multi-View Clustering Based on Anchor Graph Tensor Factorization1
Non-negative tensor factorization is used to decompose an anchor graph tensor that combines anchor graphs from multiple views to consider inter-view information comprehensively and enhances the interpretability of the factorization.
- KaLM: Knowledge-aligned Autoregressive Language Modeling via Dual-view Knowledge Graph Contrastive Learning1
An approach which fine-tunes autoregressive LLMs to align with KG knowledge via the joint objective of explicit knowledge alignment and implicit knowledge alignment, which achieves a significant performance boost in evaluations of knowledge-driven tasks, specifically embedding-based knowledge graph completion and generation-based knowledge graph question answering.
- 1
- 1
- Joint-based multi-task sparse learning for human action recognition1
A novel human action recognition method based on human joint position available video sequences that uses the joint-based covariance matrix over time as a discriminative descriptor for each human joint to encode the temporal information of the moving joints.
- Linear Regression Fisher Discrimination Dictionary Learning for Hyperspectral Image Classification1
The proposed method, linear regression Fisher discrimination dictionary learning (LRFDDL), obtains a more discriminative dictionary and a classifier by incorporating linear regression term and the Fisher discrimination into the objective function during training.
- 1
- Robust Image Watermarking Based on Feature Regions1
This chapter presents and discusses the framework of the extraction and selection of the scale-space feature points, and proposes two robust image watermarking algorithms through synchronizing water marking with the invariant local feature regions centered at feature points.
- –
- –
- –
- Streaming Video Moment Retrieval via Temporal Forecast Diffusion Model–
The proposed DiffuStream leverages a Temporal Forecast Diffusion Model to efficiently synthesize future semantics conditioned on observable contexts, thereby providing additional future information to assist in reasoning about the current frame.
- –
- –
- E 2 SGNN: Reconciling Expression and Efficiency in Spiking Graph Neural Network–
E2SGNN, a novel network comprising a dual-scale modulated spiking backbone and a latency-dynamic optimization module, which both breaks through binary spiking signals' expressive limitations and improves the content expressiveness of spiking graph representations, while retaining low-latency and addition-only efficient advantages.
- –
- –
- –
- –
- –
- –
- In-context Learning Demonstration Generation with Text Distillation–
This work proposes to generate representative demonstrations through a Distillation-based Demonstration Generation (DDG) framework and designs a generative model and tries to refine it by minimizing the discrepancies between the calculative models trained on generated demonstrations and the original datasets respectively.
- –
- –
- Taming Transformer for Emotion-Controllable Talking Face Generation–
This paper employs two pre-training strategies to disentangle audio into independent components and quantize videos into combinations of visual tokens and proposes the emotion-anchor (EA) representation that integrates the emotional information into visual tokens.
- A Tale of Two Experts: Cooperative Learning for Source-Free Unsupervised Domain Adaptation–
The Experts Cooperative Learning (EXCL) is proposed, a three-stage pipeline that collaboratively retrieves pseudo-source and complex target samples, and separately fine-tunes each expert on its respective sample set, and enforces learning object consistency via a shared learning result.
- –
- –
- –
- [Prediction of protein Kbhb sites based on learnable feature embedding].–
Experimental results showed that the proposed feature embedding learning method, AutoTF-Kbhb, can be used as an auxiliary means to identify Kbhb sites, significantly outperforming other methods in comparison.
- Bi-Orthogonal Non-negative Tensor tri-Factorization for Tensorized Label Learning–
This paper studies bi-orthogonal constraint due to the fact that it leads to rigorous interpretability of clustering, applies it to multimedia data label learning and produces a novel co-multi-view label learning based on bi-orthogonal 3-factor NTF.
- –
- –
- Readability Trends in Four Genres: A Research Based on COCA–
It is concluded that there may be multiple factors at work behind the diachronic development of readability and that these factors may have different effects on different genres.
- –
- –
- One-Step Multi-View Clustering Based on Transition Probability–
This work introduces the One-Step Multi-View Clustering Based on Transition Probability (OSMVC-TP), a probabilistic approach, which leverages the anchor graph, representing the transition probabilities from samples to anchor points, and calculates the transition probabilities from samples to categories, enhancing the interpretability of clustering.
- Development of mechanical design and manufacturing based on KNN algorithm–
In order to better meet the needs of China's social development, the original form of economic development must constantly change, update the original development concept, increase the innovation of mechanical design and manufacturing industry, so that it can better meetThe needs of social development and improve its progressiveness level.
- –
- –
- Deep Bayesian Quantization for Supervised Neuroimage Search–
Deep Bayesian Quantization seamlessly combines the deep representation learning and the representation compact quantization within a novel Bayesian learning framework, where a proxy embedding-based likelihood function is developed to alleviate the sampling issue for traditional similarity supervision.
- –
- –
- –
- –
- –
- –
- Special issue on selected and extended papers from the 2015 International Conference on Intelligence Science and Big Data Engineering (IScIDE 2015)–
A synopsis of papers collected in the special issue of IScIDE 2015 are presented, which address some aspects of the above reearch problems.
- –
- –
- Recognizing Human Actions by Sharing Knowledge in Implicit Action Groups–
A multi-task learning framework with group-structured regularization to share knowledge in category groups to encourage the categories in the same group to share dimensions feature from the same Gaussian component and vice versa.
- –
- Teaching Systems for Excellent Course of Digital Circurt and System Design–
The practice shows that three teaching systems can meet different students' requirement and are efficient to excite the students' study interesting and to improve the comprehensive ability of the students.
- Robust image watermarking scheme using local feature regions and tchebichef moments–
A robust image watermarking scheme based on local Tchebichef moments that is imperceptible and robust against common image processing operations and geometric distortions, even combined attacks is proposed.
- –
- –
- –
- Method to Measure the Active Distance of RFID Tag–
The system principle is explained in detail and which factors affected the active distance are deduced and the paper is to the HRF 13.
- –
- Practice and Experience of EDA and the Fundamentals of ASIC Design in the Major of Electronic and Electric Information–
The importance of the courses about EDA and the fundamentals of ASIC design in the major of electronics and electrics and a synchronous constructive project is presented.
- –
- Application of three\|dimension animation in the simulation of wing\|beat patterns of insects.–
The wing\|beat patterns of over sixty species of tethered insects were observed and they could be divided into three kinds and nine types according to the movement of wings.
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.