Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Edwin R. HancockSuggested from co-authorship
- Richard C. WilsonSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works221 from public data
- Binary neural networks: A survey597
A comprehensive survey of algorithms proposed for binary neural networks, mainly categorized into the native solutions directly conducting binarization, and the optimized ones using techniques like minimizing the quantization error, improving the network loss function, and reducing the gradient error are presented.
- DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalization383
DNGaussian is introduced, a depth-regularized framework based on 3D Gaussian radiance fields, offering real-time and high-quality few-shot novel view synthesis at low costs, and a Hard and Soft Depth Regularization to restore accurate scene geometry under coarse monocular depth supervision while maintaining a fine-grained color appearance.
- 308
- Self-Trained Deep Ordinal Regression for End-to-End Video Anomaly Detection257
This work devise an end-to-end trainable video anomaly detection approach that enables joint representation learning and anomaly scoring without manually labeled normal/abnormal data and outperforms state-of-the-art methods that require no labeled training data.
- 256
- Beyond Triplet Loss: Person Re-Identification With Fine-Grained Difference-Aware Pairwise Loss230
This work introduces a novel pairwise loss function that enables ReID models to learn the fine-grained features by adaptively enforcing an exponential penalization on the images of small differences and a bounded penalization of large differences.
- Deep learning-based 3D point cloud classification: A systematic survey and outlook217
The purpose of this paper is to provide researchers in this field with the latest research progress and future trends, and to summarize deep learning-based methods for point cloud classification and complement recent research work.
- Adaptive Unimodal Cost Volume Filtering for Deep Stereo Matching212
This paper proposes to directly add constraints to the cost volume by filtering cost volume with unimodal distribution peaked at true disparities and achieves state-of-the-art performance on Scene Flow and two KITTI stereo benchmarks.
- 198
- 191
- Goal-Oriented Gaze Estimation for Zero-Shot Learning171
This work introduces a novel goal-oriented gaze estimation module (GEM) to improve the discriminative attribute localization based on the class-level attributes for ZSL and implies the promising benefits of collecting human gaze dataset and automatic gaze estimation algorithms on high-level computer vision tasks.
- Graph characteristics from the heat kernel trace159
This paper explores how permutation invariants computed from the trace of the heat kernel can be used to characterize graphs for the purposes of measuring similarity and clustering, and compares them with the use of the Laplacian spectrum.
- 148
- 141
- 137
- 127
- 127
- 126
- 124
- 124
- 121
- 98
- SHREC'12 Track: Generic 3D Shape Retrieval96
The aim of this track is to measure and compare the performance of generic 3D shape retrieval methods implemented by different participants over the world and their retrieval accuracies were evaluated using 7 commonly used performance metrics.
- Material based salient object detection from hyperspectral images93
A material-based salient object detection method which can effectively distinguish objects with similar perceived color but different spectral responses, and outperforms several existing hyperspectral salient object Detection approaches and the state-of-the-art methods proposed for RGB images.
- Revisiting Domain Generalized Stereo Matching Networks from a Feature Consistency Perspective91
A simple pixel-wise contrastive learning across the viewpoints for stereo matching performance to unseen domains is proposed, which achieves superior performance over several state-of-the-art networks.
- Adaptive hash retrieval with kernel based similarity87
A novel adaptive similarity measure which is consistent with k-nearest neighbor search is presented, and it is proved that it leads to a valid kernel if the original similarity function is a kernel function.
- Unsupervised feature selection via adaptive autoencoder with redundancy control82
A novel adaptive autoencoder with redundancy control (AARC) as an unsupervised feature selector is presented, adding two Group Lasso penalties to the objective function and comparisons with nine state-of-the-art methods illustrate the efficiency of AARC for un supervised feature selection.
- 79
- Autonomous detection of damage to multiple steel surfaces from 360° panoramas using deep neural networks75
A new PAnoramic surface damage DEtection Network (PADENet) is presented, employing the proposed multiple projection methods to process high‐resolution images, and modifying the faster region‐based convolutional neural network and training via transfer learning on VGG‐16, which improves the precision for detecting multiple types of damage in distortion.
- Latent Distribution Preserving Deep Subspace Clustering73
A novel deep subspace clustering method based on a latent distribution-preserving autoencoder is proposed, which introduces a distribution consistency loss to guide the learning of distribution- Preserving latent representation, and consequently enables strong capacity of characterizing the real-world data for sub space clustering.
- 70
- 70
- 69
- Learning binary code for fast nearest subspace search68
A new approach for hashing-based ANS search which can directly binarize a subspace without transforming it into a vector, and simultaneously leverages the learned binary codes for subspaces to train matrix classifiers as hash functions.
- Optimized graph-based segmentation for ultrasound images68
A novel technique named as parameter-automatically optimized robust graph-based (PAORGB) image segmentation method is accordingly proposed and performed on breast ultrasound images to extract breast tumors in ultrasound images more adaptively and accurately.
- Learning Transferable Negative Prompts for Out-of-Distribution Detection66
A novel OOD detection method, named ‘NegPrompt’, to learn a set of negative prompts, each representing a negative connotation of a given class label, for delineating the boundaries between ID and OOD images.
- 66
- 61
- 61
- 60
- Geometric characterization and clustering of graphs using heat kernel embeddings59
This paper investigates the use of heat kernels as a means of embedding the individual nodes of a graph in a vector space and investigates several alternative methods of characterization, including spatial moments for the embedded points, the Laplacian spectrum for the Euclidean distance matrix and scalar curvatures computed from the difference in geodesic and Euclidan distances.
- 58
- 57
- 57
- Self-Supervised deep homography estimation with invertibility constraints53
A novel homography estimation approach, named “Self-Supervised Regression Network(SSR-Net)”, which relaxes the need of ground truth annotations and takes advantage of invertibility constraints is proposed, which outperforms several state-of-the-art approaches.
- 52
- 52
- 48
- Subspace Structure Regularized Nonnegative Matrix Factorization for Hyperspectral Unmixing47
A new unmixing method is proposed that learns a subspace structure from the original image and incorporates it into the sparse NMF framework to promote un Mixing performance.
- 45
- 45
- 44
- A generative model for graph matching and embedding44
This paper shows how to construct a generative model for graph structure through the embedding of the nodes of the graph in a vector space by performing a Young-Householder decomposition of the heat-kernel into an inner product of node co-ordinate matrices.
- 41
- Select, Extract and Generate: Neural Keyphrase Generation with Layer-wise Coverage Attention41
SEG-Net is proposed, a neural keyphrase generation model that is composed of a selector that selects the salient sentences in a document and an extractor-generator that jointly extracts and generates keyphrases from the selected sentences.
- Cohesion Intensive Deep Hashing for Remote Sensing Image Retrieval41
A cohesion intensive deep hashing model for remote sensing image retrieval that is terminated by a Heaviside-like function for binarizing remote sensing images and adopted a gradualness optimization method for obtaining optimal model parameters in order to favor accurate binary codes with little quantization error.
- 40
- A Hierarchical Convolution Neural Network (CNN)-Based Ship Target Detection Method in Spaceborne SAR Imagery40
A ship target detection method in spaceborne SAR imagery, using a hierarchical convolutional neural network (H-CNN), which can outperform the conventional constant false-alarm rate technique and CNN-based models.
- Privacy-preserved community discovery in online social networks40
A novel framework is proposed to discover community structure in online social networks while preserving sensitive link information and outperforms the baseline algorithm and can achieve a high accuracy on the discovered community structure.
- 40
- 39
- 39
- 38
- 38
- Heat Kernels, Manifolds and Graph Embedding38
This paper investigates the use of heat kernels as a means of embedding graphs in a pattern space using multidimensional scaling and uses the eigenvalues of the Laplacian matrix to characterise the distribution of points representing the embedded nodes.
- Multi-head enhanced self-attention network for novelty detection37
This work proposes multihead enhanced self-attention in the AlOCC network, thereby increasing the difference between the IC and OC samples and significantly increasing OCC accuracy compared with ALOCC accuracy, and proposes a new loss, called adversarial-balance loss, that effectively solves the training instability problem, further increasing O CC accuracy.
- Deep Residual Convolutional Neural Network for Hyperspectral Image Super-Resolution37
A deep residual convolutional neural network is proposed to increase the spatial resolution of hyperspectral image by minimizing the difference between the estimated image and the ground truth high resolution image.
- Quantum kernels for unattributed graphs using discrete-time quantum walks37
A new family of graph kernels where the graph structure is probed by means of a discrete-time quantum walk and the kernel is defined as the negative exponential of the quantum JensenShannon divergence between their density matrices.
- 33
- Exploring the Usage of Pre-trained Features for Stereo Matching29
This paper develops the deformable neck (DN) that decouples the network architectures of pre-trained backbones and stereo matching networks, and reveals the crucial factors that influence using pre-trained features for stereo matching.
- 29
- 29
- Cross-modal hashing with semantic deep embedding29
This paper introduces a novel end-to-end deep cross-modal hashing framework which integrates feature and hash-code learning into the same network and proposes a novel network structure and a loss function with dual semantic supervision for hash learning.
- Deep Hashing by Discriminating Hard Examples28
A novel Gamma distribution-enabled and symmetric Kullback-Leibler divergence-based loss is introduced, which works similarly as imposing two smoothed hinge losses on the respective similar and dissimilar pairs, which enables the model to well encode the key information carried by both easy and hard examples.
- 27
- 27
- 26
- 26
- 24
- 24
- Learning invariant structure for object identification by using graph methods24
This paper describes a classifier which is unique in being able to learn class identity no matter how the class instances are depicted, and depends on spectral graph analysis of a hierarchical description of an image to construct a feature vector of fixed dimension.
- 23
- 22
- 21
- A graph kernel based on the Jensen-Shannon representation alignment21
By aligning JS representations of vertices, this paper identifies the correspondence between the vertices of two graphs and this allows us to construct a matching-based graph kernel that addresses the drawback of neglecting the relative locations between substructures that arises in the R-convolution kernels.
- 20
- 20
- A Graph Kernel from the Depth-Based Representation20
A novel graph kernel is developed by matching the depth-based substructures in graphs by establishing a possible correspondence between vertices of two graphs that allows us to construct a matching-based graph kernel.
- 19
- 19
- Learning Binary Codes with Bagging PCA19
This paper attempts to leverage the bootstrap sampling idea and integrate it with PCA, resulting in a new projection method called BaggingPCA, in order to learn effective binary codes, closely connected with the core idea of LSH.
- 18
- 18
- 18
- SHREC'12 Track: Stability on Abstract Shapes18
This contribution reports the results of the SHREC 2012 track: Stability on Abstract Shapes, which saw six registrations of which only three participants effectively sent theresults of their runs.
- 17
- 16
- Graph Matching using Spectral Embedding and Semidefinite Programming16
This paper uses semidefinite programming to develop a variant of the Scott and Longuet-Higgins algorithm to find point correspondences and experiment with the resulting algorithm on a number of real-world problems.
- 15
- Deep supervised hashing using symmetric relative entropy15
This work proposes a novel deep supervised hashing method that aims to minimize the information loss generated during the embedding process, measured by the Jensen–Shannon divergence to ensure that compact hash codes have a similar distribution with those from the original images.
- Deep depth-based representations of graphs through deep learning networks15
It is theoretically demonstrate that the deep depth-based representations of graphs not only reflect both the local and global characteristics of graphs through the depth- based representations, but also capture the main structural relationship and information content over all graphs under investigations.
- 15
- Information bottleneck and selective noise supervision for zero-shot learning14
This work proposes a novel bidirectional embedding based generative model with a tight visual-semantic coupling constraint that outperforms the state-of-the-art methods in different ZSL settings on most benchmark datasets.
- 14
- 14
- OpenCIL: Benchmarking Out-of-Distribution Detection in Class-Incremental Learning13
A new baseline for OOD detection in CIL is proposed, namely Bi-directional Energy Regularization ($\textbf{BER}$), which is specially designed to mitigate these two biases in different CIL models by having energy regularization on both old and new classes.
- Discriminative features matter: Multi-layer bilinear pooling for camera localization13
A new network for camera localization using the proposed feature pooling module for feature aggregation, which shows that focusing on discriminative features significantly improves the network performance of camera localization in most cases.
- 13
- A novel logistic multi-class supervised classification model based on multi-fractal spectrum parameters for hyperspectral data13
The proposed approach achieves better accuracy with lower computational cost simultaneously and is compared with the logistic regression classification model based on an adaptive bands selection method by using the Airborne Visible/Infrared Imaging Spectrometer and airborne Push Hyperspectral Imager data.
- 12
- 12
- 12
- Long-Tailed Out-of-Distribution Detection via Normalized Outlier Distribution Adaptation12
To obtain a more reliable set of predicted OOD samples on long-tailed ID data, a novel dual-normalized energy loss is introduced in AdaptOD, which leverages class- and sample-wise normalized energy to enforce a more balanced prediction energy on imbalanced ID samples.
- Binary Coding by Matrix Classifier for Efficient Subspace Retrieval12
A new approach for hashing-based ANS retrieval that learns the binary codes for given subspace set following a similarity preserving criterion and simultaneously leverages the learned binary codes to train matrix classifiers as hash functions to efficiently solve the large-scale and high-dimensional multimedia data retrieval problem.
- P2P Lending Analysis Using the Most Relevant Graph-Based Features12
This work develops a novel filter-based feature selection method based on graph-based features that incorporates the relationships between pairwise feature samples into the feature selection process and select an optimal subset of features based on the most relevant graph- based features, through the Jensen-Shannon divergence measure.
- A Spectral Generative Model for Graph Structure12
It is shown how to use this model to both project individual graphs into the eigenspace of the point-position covariance matrix and how to fit the model to potentially noisy graphs to reconstruct the Laplacian matrix.
- Asymmetric propagation based batch mode active learning for image retrieval11
A novel batch mode active learning scheme for informative sample selection inspired by the method of graph propagation is presented, which takes the correlation between labeled samples and unlabeled samples, but the correlation among unlabeling samples taken into account as well.
- Isotree: Tree clustering via metric embedding11
This paper uses the ISOMAP algorithm to embed the trees in a Euclidean space using the pattern of shortest distances between nodes, and uses the spectrum of the Laplacian matrix for the embedded graphs to lift the co-spectrality of the trees.
- Manifold Embedding of Graphs Using the Heat Kernel11
This paper investigates the use of heat kernels as a means of embedding the individual nodes of a graph on a manifold in a vector space by exponentiating the Laplacian eigen-system over time, and shows how the spectral representation of the heat kernel can be used to compute both Euclidean and geodesic distances between nodes.
- 10
- Local-global nested graph kernels using nested complexity traces10
Experiments on standard graph datasets abstracted from bioinformatics and computer vision databases demonstrate the effectiveness and efficiency of the proposed graph kernels, namely the nested aligned kernel and the nested reproducing kernel, drawing on depth-based complexity traces.
- 10
- An incremental structured part model for object recognition10
The method captures the inherent connections of the semantic parts of objects and characterizes the structural relationships between them and demonstrates advantages over several state-of-the-art part-based methods in the literature.
- Cache refreshing for online social news feeds10
This work proposes a novel cache scheme called SOCR (Social Online Cache Refreshing) for identifying and refreshing cache entries, and uses a utility-based strategy to accurately identify cache entries that need to be refreshed.
- 10
- Toward personalized peer-to-peer top-k processing10
This work presents the first personalized peer-to-peer top-k search protocol for a collaborative tagging system, where each peer maintains relevant personalized information about its tagging behavior as well as that of its social neighbors, and uses those to locally process its queries.
- 9
- SCCNet: Self-correction boundary preservation with a dynamic class prior filter for high-variability ultrasound image segmentation9
A two-stage network using a self-correction boundary preservation module and class-context filter to alleviate problems in ultrasound image segmentation tasks, showing at least a 1.2-3.7% improvement on the three datasets, Neck Muscle, Thyroid, and CAMUS.
- 9
- Pattern recognition for high performance imaging9
The goal of this special issue is to provide a forum for researchers and practitioners in the broad computer vision and pattern recognition community to present their novel and original pattern recognition research for high performance imaging.
- Manifold embedding for shape analysis9
This paper describes how graph-spectral methods can be used to transform the node correspondence problem into one of point-sets alignment, and constructs a statistical point distribution model for the embedded node positions using the eigenvalues and eigenvectors of the covariance matrix.
- 8
- Maximum margin hashing with supervised information8
Inspired by support vector machine which achieves strong generalization capability by maximizing the margin of its decision surface, a binary hash function is proposed in the same manner and can achieve better performance than the state-of-the-art hashing methods.
- 8
- Trace Formula Analysis of Graphs8
The use of the heat-kernel moments as a means of characterising graph structure for the purposes of clustering and experiments with the COIL and Oxford-Caltech databases reveal the effectiveness of the representation.
- 8
- 7
- 7
- 6
- Towards Fully Decoupled End-to-End Person Search5
A task-incremental person search network is proposed to incrementally construct an end-to-end model for the detection and re-id sub-task, which decouples the model architecture for the two sub-tasks.
- Iterative Deep Subspace Clustering5
An iterative framework for unsupervised deep subspace clustering, which first cluster the given data to update the subspace ids, and then update the representation parameters of a Convolutional Neural Network with the clustering result.
- Deep Homography Estimation with Pairwise Invertibility Constraint5
This paper proposes a Siamese network with pairwise invertibility constraint for supervised homography estimation and adopts the matrix representation of the homography rather than the commonly used 4-point parameterization in other methods.
- Bilinear Discriminant Analysis Hashing: A Supervised Hashing Approach for High-Dimensional Data5
The results show that the proposed supervised hashing method Bilinear Discriminant Analysis Hashing can achieve comparable accuracy to the state-of-the-art supervised hashing schemes, while, however, cost much less time and memory.
- Discriminative sparse neighbor coding5
This work establishes two modules to improve the discriminative ability of sparse representation, and incorporates the feature distribution into the objective function, spanning a class-specific low dimensional subspace for effective sparse coding.
- Describing and learning of related parts based on latent structural model in big data5
A novel latent structural model for big data image recognition that considers not only independent components but also their inherent spatial and appearance relationships and has achieved better performance than several state-of-the-art methods.
- Salient Region Detection for Biometric Watermarking5
This chapter presents two salient-region-detection based biometric watermarking scenarios, in which robust annotation and fragile authentication watermark are respectively applied to biometric systems.
- Autofocusing of SAR images using STFRFT-based preprocessing5
This work proposes an advanced PPP algorithm with a preprocessing technique based on the short-time fractional Fourier transform (STFRFT) to relax the limitation on the scene content.
- 5
- Recent Results on Heat Kernel Embedding of Graphs5
This paper describes how heat-kernel asymptotics can be used to compute approximate Euclidean distances between nodes in a graph by performing Multidimensional Scaling (MDS).
- Graph Matching Using Manifold Embedding5
This paper describes how graph-spectral methods can be used to transform the node correspondence problem into one of point-set alignment, and uses a variant of the Scott and Longuet-Higgins algorithm to find point correspondences.
- 4
- 4
- 4
- 4
- Hierarchical Hashing for Image Retrieval4
This method models and fuses both hierarchical semantic level relationship through taxonomy structure of dataset and feature level relationship of images into an integrated learning objective, then an optimization scheme is developed to solve the learning problem.
- Hypergraph Spectra for Semi-supervised Feature Selection4
A novel hypergraph based semi-supervised feature selection algorithm to select relevant features using both labeled and unlabeled data is proposed and a hypergraph subspace learning view of feature selection is derived which casting the feature discriminant analysis into a regression framework that considers the correlations among features.
- 4
- 3
- 3
- HMFlow: Hybrid Matching Optical Flow Network for Small and Fast-Moving Objects3
This work proposes a new Hybrid Matching Optical Flow Network (HMFlow) by integrating GMC into existing coarse-to-fine networks seamlessly, which can apply global matching features to guide the network to discover the small and fast-moving objects mismatched by local matching features.
- Select, Extract and Generate: Neural Keyphrase Generation with Syntactic Guidance.3
SEG-Net is proposed, a neural keyphrase generation model that is composed of two major components, a selector that selects the salient sentences in a document, and an extractor-generator that jointly extracts and generates keyphrases from the selected sentences.
- Differential weights-based band selection for hyperspectral image classification3
A supervised hyperspectral band selection method based on differential weights, which depict the contribution degree of each band for classification is proposed, which helps to reduce the expensive cost of computation and storage.
- Personalized top-k processing: from centralized to decentralized systems3
Experimental results on real datasets show that on-line personalization is promising to fulfill the diverse user preferences while the proposed algorithms make it feasible in both centralized and decentralized systems.
- 2
- Predicting Positive Repeat Prostate Biopsy Outcomes: Comparison of Machine Learning Approaches to Identify Key Parameters and Optimal Algorithms2
Initial and latest prostate volumes, initial and latest PSA levels, latest fPSA/PSA ratio and age are significant factors for this model, and the variable ranking emerged as the most effective method for identifying the essential factors influencing repeat biopsy results.
- 2
- Fast Subspace Clustering Based on the Kronecker Product2
A subspace clustering model based on the Kronecker product that significantly improves the efficiency compared with several state-of-the-art methods and reduces the computational complexity to O(kN3/k).
- A One-step Pruning-recovery Framework for Acceleration of Convolutional Neural Networks2
This paper accelerates the convolutional neural network in one-step pruning-recovery manner with a novel optimization objective function, which achieves higher accuracy with much less cost compared with existing pruning methods.
- 2
- Deep Supervised Hashing with Information Loss2
This work proposes a novel deep supervised hashing method which aims to minimize the information loss during low-dimensional embedding process, and uses Kullback-Leibler divergences to constrain the compact codes having a similar distribution with the original images.
- Isometric Mapping Hashing2
This paper presents a hashing scheme that preserves the geodesic distance and use a feasible out-of-sample method to generate the binary codes efficiently and shows that this method outperforms several alternative hashing methods.
- Learning Graph Model for Different Dimensions Image Matching2
This paper addresses the problem of matching hyperspectral images with different dimensions by investigating structured support vector machine to learn graph model for each type of image, which incorporates both low-level features and stable correspondences within images.
- 2
- Design and Implementation of Visible Human-Machine Interface for Trajectory Tracking in Agriculture Vehicle Navigation2
The design of map views including field view and machine view for field computer system based on coordinate conversion is introduced and given a bird's eye view of the map.
- 2
- Shape Decomposition for Graph Representation2
A new method based on a refined morphological shape decomposition process that can solve two problems for morphological decomposition - scale invariant and noise.
- 1
- Revisiting Continual Ultra-fine-grained Visual Recognition with Pre-trained Models1
This work establishes the first comprehensive C-UFG benchmark using massive soy leaf data and proposes the first unified model, Unic, that is capable of tackling the C-UFG problem where VC-UFG and HC-UFG co-exist in a single continual learning sequence.
- 1
- 1
- 1
- 1
- 1
- 1
- Cross-Model Retrieval with Reconstruct Hashing1
This paper proposes a novel end-to-end deep cross-modal hashing framework, integrating feature and hash-code learning into the same network, keeping the relationship of features between modalities.
- 1
- 1
- A Novel Image Segmentation Algorithm Based on Multiple Features Fusion with Hypergraph and Superpixel1
This paper presents an image segmentation method which utilizes image multiple features fusion, superpixel segmentation technique and hypergraph theory to form a graph structure which can be represented as an affinity matrix.
- 1
- An Incremental Structured Part Model for Image Classification1
A novel incremental method is presented, which learns a part model to classify objects using only a small number of training samples and captures the inherent connections of the semantic parts of objects and builds structural relationship between them.
- The Innovative Design System of Traditional Embroidery Patterns Based on Computer Linear Classifier Intelligent Algorithm Model–
An innovative system for designing traditional embroidery patterns utilizing a computer-based linear classifier intelligent algorithm achieves efficient classification and recognition of embroidery pattern features by employing the Fisher linear discriminant analysis technique, thus enabling the intelligent and innovative creation of designs.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Accelerating Deep Convnets via Sparse Subspace Clustering–
This paper proposes a novel filter pruning method that reduces the redundancy in one convolutional layer by applying sparse subspace clustering to its output feature maps and achieves state-of-the-art results after fine-tuning.
- –
- –
- Learning Deep Embeddings via Margin-Based Discriminate Loss–
This paper proposes a new metric learning objective called Margin-based Discriminate Loss which tries to keep the similar and the dissimilar strictly discriminate by pulling multiple positive examples together while pushing multiple negative examples away at each update.
- –
- –
- A Large Margin Learning Method for Matching Images of Natural Objects with Different Dimensions–
This chapter addresses the problem of matching hyperspectral image with other types of images by investigating structured support vector machines to construct and learn a graph-based model for each type of image.
- –
- –
- Discriminative Feature Selection in Image Classification and Retrieval–
A new method is presented to improve the performance of current bag-of-words based image classification process by introducing a pairwise image matching scheme to select the discriminative features.
- A New Multi-Focus Image Fusion Algorithm Based on Contrast Ratio and Discrete Wavelet Frame Transform–
A multi-focus image fusion algorithm based on contrast ratio and discrete wavelet frame transform and evaluation indicators of information entropy, standard deviation, average gradient and spatial frequency objectively evaluates the fusion quality of this algorithm.
- –
- –
- –
- –
- –
- –
- –
- Quantitative Evaluation on Heat Kernel Permutation Invariants–
This paper first shows how permutation invariants computed from the trace of the heat kernel can be used to characterize graphs for the purposes of measuring similarity and clustering, and explores three different approaches to characterize the heat Kernel trace as a function of time.
- –
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.