Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks327 from public data
- 2,901
- 1,750
- The Sixth Visual Object Tracking VOT2018 Challenge Results804
The Visual Object Tracking challenge VOT2018 is the sixth annual tracker benchmarking activity organized by the VOT initiative; results of over eighty trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in the recent years.
- 722
- 268
- Pansharpening via Detail Injection Based Convolutional Neural Networks254
This paper designs a new detail injection based convolutional neural network (DiCNN) framework for pansharpening with the MS details being directly formulated in end-to-end manners, where the first detail injections based CNN mines MS details through the PAN image and the MS image, and the second one utilizes only the PAN picture.
- A Real-Time Cross-Modality Correlation Filtering Method for Referring Expression Comprehension253
A novel Realtime Cross-modality Correlation Filtering method (RCCF) that reformulates the referring expression comprehension as a correlation ï¬ltering process and achieves leading performance in RefClef, RefCOCO, Ref COCO+ and RefC OCOg benchmarks.
- Referring Image Segmentation via Cross-Modal Progressive Comprehension250
A Cross-Modal Progressive Comprehension (CMPC) module and a Text-Guided Feature Exchange (TGFE) module to effectively address the challenging task of referring image segmentation.
- 207
- 190
- SemanticAdv: Generating Adversarial Examples via Attribute-Conditioned Image Editing150
This paper proposes SemanticAdv to generate a new type of semantically realistic adversarial examples via attribute-conditioned image editing and demonstrates the applicability of SemanticAdv on both face recognition and general street-view images to show its generalization.
- Change detection from synthetic aperture radar images based on neighborhood-based ratio and extreme learning machine145
The experimental results obtained show that the proposed SAR image change detection method is robust to speckle noise and is effective to detect change information among multitemporal SAR images.
- 144
- 134
- A Surface Defect Detection Method Based on Positive Samples133
Experiments show that the proposed GAN+LBP algorithm and supervised training algorithm with sufficient training samples have fairly high detection accuracy and because of its unsupervised characteristics, it has higher practical application value.
- TrojDiff: Trojan Attacks on Diffusion Models with Diverse Targets125
An effective Trojan attack against diffusion models, TrojDiff, which optimizes the Trojan diffusion and generative processes during training and achieves high attack performance under different adversarial targets using different types of triggers, while the performance in benign environments is preserved.
- 121
- 107
- 96
- Multi-attributed heterogeneous graph convolutional network for bot detection95
A novel bot detection framework, namely Bot-AHGCN, is proposed, which models fine-grained network flow objects as a multi-attributed heterogeneous graph and transforms bot detection problem into a semi-supervised node classification task on the graph.
- 88
- Too Far to See? Not Really!—Pedestrian Detection With Scale-Aware Localization Policy88
An active pedestrian detector that explicitly operates over multiple-layer neuronal representations of the input still image and is demonstrated to produce overall lower detection errors on widely used benchmarks, and it works particularly well with far-scale pedestrians.
- 88
- MME-Survey: A Comprehensive Survey on Evaluation of Multimodal LLMs85
This work aims to offer researchers an easy grasp of how to effectively evaluate MLLMs according to different needs and to inspire better evaluation methods, thereby driving the progress of MLLM research.
- 83
- Integrating Context and Occlusion for Car Detection by Hierarchical And-Or Model81
A method of learning reconfigurable hierarchical And-Or models to integrate context and occlusion for car detection using Weak-Label Structural SVM and compares with state-of-the-art variants of deformable part-based models and other methods.
- 77
- 76
- Recent advances in pansharpening and key problems in applications76
This article provides a review on recent advances in solving or reducing the problems of colour distortion and spatial artefacts and provides experimental results to show the capacity of various pansharpening techniques in handling the key problems.
- 75
- 75
- A fast face detection method via convolutional neural network73
Experimental results on several popular face detection datasets show the efficiency and the effectiveness of the proposed fast face detection method based on discriminative complete features extracted by an elaborately designed convolutional neural network.
- A fast image dehazing algorithm based on negative correction70
The concept of negative correction inspired by the practical application of photographic developing is introduced and a fast image dehazing algorithm is accordingly proposed that can effectively remove hazes and significantly reduce the computational complexity.
- Naive Gabor Networks for Hyperspectral Image Classification67
The naive Gabor networks or Gabor-Nets are introduced that, for the first time in the literature, design and learn CNN kernels strictly in the form of Gabor filters, aiming to reduce the number of involved parameters and constrain the solution space and, hence, improve the performances of CNNs.
- 66
- 65
- 64
- Single image haze removal using content‐adaptive dark channel and post enhancement63
This study proposes a method of single image haze removal using content-adaptive dark channel and post enhancement, and demonstrates that the proposed method significantly improves the visibility of the hazy image.
- Sparse Black-Box Video Attack with Reinforcement Learning59
This work formulate the black-box video attacks into a Reinforcement Learning (RL) framework and demonstrates that the proposed method can significantly reduce the adversarial perturbations with efficient query times.
- 54
- 53
- An Imperceptible Medical Image Watermarking Framework for Automated Diagnosis of Retinal Pathologies in an eHealth Arrangement52
An imperceptible watermarking-based security framework to address the issues of authentication and integrity verification of medical images for eHealth applications and validate that the insertion of the watermark in the retinal images does not affect the automated medical image diagnosis of various retinal pathologies.
- 50
- 50
- 50
- SuperVAE: Superpixelwise Variational Autoencoder for Salient Object Detection48
This paper proposes a novel salient object detection framework using a superpixelwise variational autoencoder (SuperVAE) network and proposes a perceptual loss to take advantage from deep pre-trained CNNs to train the SuperVAE network.
- 48
- 48
- 48
- 47
- Region-Based Rate Control for H.264/AVC for Low Bit-Rate Applications47
Experimental results demonstrate that the proposed region-based rate control approach can achieve both better subjective and objective quality by performing the rate control adaptively with the content, compared to the conventional rate control approaches.
- A LSTM-Based Anomaly Detection Model for Log Analysis45
A LSTM-based anomaly detection approach is designed and could successfully identify attacks on two widely-used datasets and outperforms three popular anomaly detection algorithms, one-class SVM, GMM and Principal Components Analysis, in terms of accuracy and efficiency.
- PSGAN++: Robust Detail-Preserving Makeup Transfer and Removal45
Experiments demonstrate that PSGAN++ not only achieves state-of-the-art results with fine makeup details even in cases of large pose/expression differences but also can perform partial or degree-controllable makeup transfer.
- 44
- A Naturalness Preserved Fast Dehazing Algorithm Using HSV Color Space44
A naturalness preserved fast dehazing algorithm using hue, saturation, and value (HSV) color space is proposed, which maintains naturalness by preserving hue and suppressing halo effects and the computational complexity has been largely reduced.
- 43
- 42
- Sketch-Based 3D Model Retrieval by Viewpoint Entropy-Based Adaptive View Clustering41
A sketch-based 3D model retrieval algorithm is proposed by utilizing viewpoint entropy-based adaptive view clustering and shape context matching and performs Fuzzy C-Means based view clusters on the sample views based on their viewpoint entropy values.
- 40
- Deep Ensemble Learning Based Objective Grading of Macular Edema by Extracting Clinically Significant Findings from Fused Retinal Imaging Modalities40
This paper proposes an automated framework for the classification of ME and healthy eyes using retinal fundus and OCT scans based on deep ensemble learning and achieved the accuracy of 94.33% for diagnosing Me and healthy subjects.
- TRS: Temporal Request Scheduling with bounded delay assurance in a green cloud data center40
This work proposes a Temporal Request Scheduling algorithm (TRS) that jointly considers the temporal diversity, and can achieve higher throughput and lower grid energy cost for a GCDC while meeting each request’s delay requirement.
- 40
- 38
- 38
- Rethinking Imitation-based Planner for Autonomous Driving37
The results demonstrate that a well-designed, purely imitation-based planner can achieve highly competitive performance compared to state-of-the-art methods involving hand-crafted rules and exhibit superior generalization capabilities in long-tail cases.
- Perceptual hash-based feature description for person re-identification37
The proposed method uses perceptual hashing to binarize low-level feature maps and combines several feature channels for feature encoding and three regional statistics are computed for hierarchical feature description.
- 36
- 35
- 35
- 34
- 33
- Automatic Ship Detection in Optical Remote Sensing Images Based on Anomaly Detection and SPP-PCANet33
The anomaly detection algorithm, based on the multivariate Gaussian distribution, regards a ship as an abnormal marine area, effectively extracting candidate regions of ships, and the combined PCANet and spatial pyramid pooling to reduce the amount of false positives and improve the detection rate is proposed.
- ESRGAN-DP: Enhanced super-resolution generative adversarial network with adaptive dual perceptual loss32
enhanced super-resolution generative adversarial network (ESRGAN) with strong learning capability is used in this paper to adapt the complexity of DP Loss, and a dynamic weighting method is proposed to eliminate the magnitude difference between perceptual losses.
- Manifold Alignment via Corresponding Projections32
Different from the previous algorithms of semi-supervised manifold alignment, this method learns the explicit corresponding projections from each original observation space to the common embedding space everywhere and could process new test data directly rather than re-alignment.
- Environmental Sound Classification Based on Multi-temporal Resolution Convolutional Neural Network Combining with Multi-level Features31
Results demonstrate that the proposed method is highly effective in the classification tasks by employing multi-temporal resolution and multi-level features, and it outperforms the previous methods which only account for single- level features.
- A Hybrid Deep Learning Framework for Integrated Segmentation and Registration: Evaluation on Longitudinal White Matter Tract Changes30
A novel hybrid convolutional neural network that integrates segmentation and registration into a single procedure is proposed that can serve as a novel tool to support clinical and epidemiological analyses on understanding microstructural brain changes over time.
- Cost-aware request routing in multi-geography cloud data centres using software-defined networking29
This work develops heuristic algorithms to solve the total cost minimisation problem and to provide a cost-aware request routing for joint optimisation of the selection of ISPs and the number of servers in distributed CDCs and proposes a regression-based workload prediction method to obtain future incoming workload.
- 28
- An Improved Water Surface Images Segmentation Algorithm Based on the Otsu Method27
Experimental results demonstrate that effectiveness of the improved Otsu method is generally better than the traditional Otsu method, and the other four commonly used threshold segmentation methods for the water surface image, which improves the segmentation accuracy of such images and reduces the segmentation error rate.
- Ship Detection in Multispectral Satellite Images Under Complex Environment27
A novel spectral-reflectance-based ship-detection method that takes advantage of employing the reflectance variance in objects between each band as additional information, and a neural network called lightweight fusion networks (LFNet) that combines the aforementioned reflectance and the color information of multispectral images to jointly verify the regions with ships.
- Methods for landslide detection based on lightweight YOLOv4 convolutional neural network26
A new algorithm for landslide detection in the plateau environment using the YOLOv4 model and the MobileNetv3 model to replace the backbone neural network CSPdarknet53 to improve the efficiency of landslide detection is proposed.
- Detecting ship targets in spaceborne infrared image based on modeling radiation anomalies26
The main innovation of the proposed method is inspired by the biological mechanism that human eyes are sensitive to the unusual and anomalous patches among complex background, and achieves a desired ship detection accuracy with higher recall than other classical ship detection methods.
- 24
- New Advances of the Research on Cloud Simulation24
The proposed latest research results can better support “cloud simulation” pattern, which users can access services of simulation resource and capability on demand anytime and anywhere through network and cloud simulation platform, to accomplish varied activities in the whole simulation life-circle.
- 23
- A person re-identification algorithm based on pyramid color topology feature23
A pyramid color topology (PCT) is proposed to extract WCTs on different scales of region-division and can achieve a full description of multi-scales topology information, and a person re-identification algorithm is implemented.
- Moving Vehicle Information Extraction from Single-Pass WorldView-2 Imagery Based on ERGAS-SNS Analysis23
The proposed method has good performance for moving vehicle detection and information extraction and improves the spatial and spectral analysis method by incorporating local ERGAS index analysis (ERGAS-SNS) to identify moving vehicles.
- 22
- A hierarchal BoW for image retrieval by enhancing feature salience22
A hierarchal BoW to represent each feature in an individual codebook for obtaining the undisturbed ranks from each feature for feature salience enhancement and a query model based on ordinary-least-squared (OLS) regression is established for rank aggregation.
- An Improved K-means Algorithm based on Mapreduce and Grid22
The improved k-mean algorithm compared to the traditional K-means clustering algorithm has high quality results, less iteration and has good stability, and the Parallelized algorithm has a very high efficiency in data processing, and hasGood scalability and speedup.
- 21
- 21
- 3D model alignment based on minimum projection area21
Experimental results demonstrate that MPA has a good performance in finding alignment axes which are parallel to the ideal canonical coordinate frame of models and aligning similar models in similar poses under different conditions such as model variations, noise, and initial poses.
- Highly Efficient Natural Image Matting20
An extremely light-weighted model is constructed, which achieves comparable performance with ~1\% parameters (344k) of large models on popular natural image matting benchmarks.
- Oil Spill Segmentation in Ship-Borne Radar Images with an Improved Active Contour Model20
The proposed improved Active Contour Model (ACM) achieves better results and higher operation efficiency than other automatic and semi-automatic methods for oil film detection in ship-borne radar images and provides a scientific basis to assess pollution scope and estimate the necessary cleaning materials during oil spills.
- RGB-D to CAD Retrieval with ObjectNN Dataset20
The evaluation results show that the RGB-D to CAD retrieval problem, while being challenging to solve due to partial and noisy 3D reconstruction, can be addressed to a good extent using deep learning techniques, particularly, convolutional neural networks trained by multi-view and 3D geometry.
- Shape-constrained multi-atlas segmentation of spleen in CT20
This work proposes to integrate a level set shape model into the traditional label fusion framework to create a shape-constrained multi-atlas segmentation framework and offers a statistically significant improvement in spleen labeling accuracy.
- Aircraft Detection from VHR Images Based on Circle‐Frequency Filter and Multilevel Features20
A novel detector for aircraft detection from very high-resolution (VHR) remote sensing images using a circle-frequency filter and a multi-level feature model to represent both local appearance and spatial layout of aircrafts is proposed.
- 19
- ORDNet: Capturing Omni-Range Dependencies for Scene Parsing19
The proposed ORDNet outperforms previous state-of-the-art methods on three scene parsing benchmarks including PASCAL Context, COCO Stuff and ADE20K, demonstrating the superiority of capturing omni-range dependencies in deep models for scene parsing task.
- Improved SIFT match for optical satellite images registration by size classification of blob-like structures19
The scale normalization and size classification significantly improve the correct match rate as well as computational cost of the existing scale-invariant feature transform operators.
- Design and implementation of a 3D ocean virtual reality and visualization engine19
A sea-land integration platform which can reproduce drifting and diffusion processes of oil spilling from sea bottom to surface and the result shows that oil particles blend with water well and the platform meets the requirement for real-time and interactive rendering.
- 18
- 18
- Embedded zerotree wavelets coding based on adaptive fuzzy clustering for image compression18
The experimental results show that exacter reconstructed values of wavelet coefficients can be obtained at low bit-rates, the subjective and objective quality of the reconstructed image is improved, and the proof of convergence of the algorithm is given.
- 17
- COVID‐19 diagnosis on CT scan images using a generative adversarial network and concatenated feature pyramid network with an attention mechanism17
The method can help clinicians build deep learning models using their private datasets to achieve automatic diagnosis of COVID-19 with a high precision and helps overcome the problem of limited training data when using deep learning methods to diagnose CO VID-19.
- 17
- Research on key technologies of complex product virtual prototype lifecycle management (CPVPLM)17
The author presents and discusses a three-dimensional (3D) view to reinforce the share and management of the CPVP lifecycle information and studies a framework to solve the lifecycle management of CPVP.
- Human-centric Relation Segmentation: Dataset and Solution16
This work introduces a new task named human-centric relation segmentation (HRS), as a fine-grained case of HOI-det, and proposes a Simultaneous Matching and Segmentation (SMS) framework as a solution to the HRS task.
- 16
- 15
- Towards a multi‐layers anomaly detection framework for analyzing network traffic15
A novel approach, named Multi‐Layers Anomaly Detection, which extracts and combines features from different network layers and outperforms the state‐of‐the‐art methods in terms of detection rate and false alarm rate is proposed.
- Revisiting the Trade-Off Between Accuracy and Robustness via Weight Distribution of Filters14
A sample-wise dynamic network architecture named Adversarial Weight-Varied Network (AW-Net), which focuses on dealing with clean and adversarial examples with a “divide and rule” weight strategy, which adaptively adjusts the network's weights based on regulation signals generated by an adversarial router, which is directly influenced by the input sample.
- 13
- An efficient approach to collaborative simulation of variable structure systems on multi-core machines13
An advanced parallel simulator under multi-core environments that adopts an efficient dynamic load-balancing method, which can migrate models among cores with very low cost and support dynamic core allocation on demand, to address evident load-imbalance problems brought by variable-structure.
- Future Manufacturing Industry with Cloud Manufacturing13
An envisioned picture of future manufacturing with CMfg which supports different advanced manufacturing modes, including intelligent manufacturing, sustainable manufacturing, agile manufacturing, and personalized social production mode was described.
- Joint global–local information pedestrian detection algorithm for outdoor video surveillance13
The experimental results demonstrate that the proposed algorithm can improve the overall recognition performance for the small-scale pedestrians and fully exploits global information to improve recognition performance.
- 13
- 13
- 13
- PAR-YOLO: a precise and real-time YOLO water surface garbage detection model12
A surface garbage detection model named Precise and Real-time YOLO (PAR-YOLO), with a focus on real-time performance and detection accuracy, and a Noise Suppression Module (NSM) and integrated it into the neck section are proposed.
- 12
- 12
- A multi-dimensional extension of the Lightweight Temporal Compression method12
This paper investigates the extension of LTC to higher dimensions, providing a formulation of the algorithm in an arbitrary vectorial space of dimension n and implementing the algorithm for the infinity and Euclidean norms, in spaces of dimension 2D-t and 3D+t.
- RETRACTED ARTICLE: Texture-guided volumetric deformation and visualization using 3D moving least squares12
This work proposes a texture-guided deformation technique for creating visualization styles through interactive manipulations of volumetric models using 3D MLS, and describes a GPU-based implementation to achieve real-time performance of the deformation techniques and the manipulation operators.
- Phase error correction for digital fringe projection profilometry12
The causes of the phase error are analyzed and the subsequent error corrections are introduced when the color-encoded digital fringe projection (CFP) technique is used to retrieve the phase value.
- A fast block-matching algorithm using smooth motion vector field adaptive search technique12
An effective half-stop criterion is proposed that can quickly stop the search process with good enough results and obtains almost the same results as the full search at very low compuration cost, and outperforms MVFAST and PMVFAST in speed and quality, which are adopted by MPEG-4.
- 11
- An Antijamming and Lightweight Ship Detector Designed for Spaceborne Optical Images11
A new satellite optical image dataset and a novel ship detector are proposed and an antijamming module is proposed to highlight the features of objects in the whole feature map, which can improve detection performance significantly.
- 11
- Image generation step by step: animation generation-image translation10
In the comparison test with WGAN, SAGAN, WGAN-GP and PG-GAN, the LGAN(SEG) model leads 64.36 and 12.28, respectively, fully proving the model’s superiority.
- SVM Regression and Its Application to Image Compression10
A new image compression algorithm which combines SVM regression with wavelet transform and an effective entropy coder based on run-length and arithmetic coding is used to encode the support vectors and weights.
- Reduced Dynamics and Control for an Autonomous Bicycle9
This paper proposes the reduced model for the full dynamics of a bicycle and analyzes its nonlinear behavior under a proportional control law for steering, finding that the steer coefficient must be negative, consistent with the human’s intuition of turning toward a fall.
- 9
- Structured fragment-based object tracking using discrimination, uniqueness, and validity selection9
The experimental results on a recent OTB 2013 tracking benchmark data set demonstrate that the proposed algorithm can achieve reliable tracking results even in the presence of significant appearance changes, partial occlusion, and similar disturbances.
- Parameter‐adaptive nighttime image enhancement with multi‐scale decomposition9
A parameter-adaptive nighttime image enhancement method with multi-scale decomposition that can suppress noises as well as improve details effectively and is suitable to present details.
- Digital image stabilization based on adaptive motion filtering with feedback correction9
Experimental results indicate that the proposed adaptive motion filtering algorithm is more fast-computing and adaptive for different moving modes of the camera, which can reserve the intentional motions and remove the jitters steadily.
- 8
- 8
- Towards a distributed local-search approach for partitioning large-scale social networks8
This paper proposes a distributed local search algorithm, named dLS, which enables quality and efficient partition of large-scale social graphs and shows that dLS outperforms two state-of-the-art algorithms in terms of partition quality and efficiency.
- Multi-Class Objects Detection Method in Remote Sensing Image Based on Direct Feedback Control for Convolutional Neural Network8
The proposed DFCCNN model is able to extract more discriminative low-level features under the guidance of the high-level information and can achieve a higher accuracy of object detection in remote sensing image with various complex background clutter.
- A region-based video de-noising algorithm based on temporal and spatial correlations8
The experimental results demonstrate that the proposed region-based video de-noising algorithm can not only efficiently remove the noise but also maintain temporal smoothness among neighboring frames.
- 8
- 8
- 8
- 7
- Three-dimensional spine reconstruction from biplane radiographs using convolutional neural networks7
A deep learning network for three-dimensional reconstruction of the spine from biplanar radiographs is developed and evaluated to indicate that the proposed method is reliable.
- BiMGCL: rumor detection via bi-directional multi-level graph contrastive learning7
A new rumor detection framework based onBiMGCL, which models each rumor propagation structure as bi-directional graphs and performs self-supervised contrastive learning based on node-level and graph-level instances and achieves superior detection performance compared against the state-of-the-art rumor detection methods.
- SphereDepth: Panorama Depth Estimation from Spherical Domain7
SphereDepth, a novel panorama depth estimation method that predicts the depth directly on the spherical mesh without projection preprocessing, is proposed and two hyper-parameters for the proposed spherical mesh processing framework are introduced to balance the inference speed and accuracy.
- Clustering of Copper Flotation Process Based on the AP-GMM Algorithm7
A clustering algorithm is proposed in this paper, which is referred to as AP-GMM, which can get the optimal number of categories and their respective probabilities without giving the number in advance and a new category matching theory is constructed.
- 7
- RETRACTED ARTICLE: A streak detection approach for comprehensive two-dimensional gas chromatography based on image analysis7
This paper proposes a new approach to detect streaks effectively based on image analysis techniques using a pseudo-log function and preprocessing methods to compress the original data and enhance the low-contrast data, and employs steerable Gaussian filtering to delineate streak regions based on the specific orientations of streaks.
- Simulation-Based Cyber-Physical Systems and Internet-of-Things7
This chapter gives a full picture of simulation- based cyber-physical systems (SB-CPS) and simulation-based internet-of-things ( SB-IoT) and the impact of SB-C CPS/SB-IeT on society and economy, people’s livelihood and national security is discussed.
- 7
- 7
- 7
- Research and Applications of Cloud Manufacturing in China7
Four typical CMfg cases which have been successfully implemented in group enterprises and mid-small enterprise clusters are described as well as cases introducing smart manufacturing into smart city and future works and development prospect for CMfg are raised.
- 7
- A region-based rate-control scheme using inter-layer information for H.264/SVC7
Experimental results demonstrate that the proposed region-based rate-control scheme for H.264/SVC can obtain better quality both subjectively and objectively by using different weighting factors for the three regions.
- Fast Adaptive Wavelet for Remote Sensing Image Compression7
Wavelet construction rules are proposed and a new biorthogonal wavelet construction model with parameters is built by using genetic algorithm and adopting energy compaction as the optimization object function to resolve the computation complexity problem of online construction.
- 7
- Poisson Tensor Completion via Nonconvex Regularization and Nonlocal Self-Similarity for Multi-dimensional Image Recovery6
A family of nonconvex functions are employed onto the singular values of all frontal slices of a tensor in the transformed domain in order to explore the global low-rankness of the underlying tensor.
- 6
- 6
- 6
- 6
- Moving object detection algorithm based on pixel spatial sample difference consensus6
A brand-new theoretical framework for foreground object detection based on the stable spatial relationship between current pixel and the randomly selected pixels in current frame is proposed and achieved promising performance after compared with several state-of-the art methods.
- Hyperspectral and panchromatic image fusion through an improved ratio enhancement6
The proposed ratio enhancement for hyperspectral (HS) image pansharpening achieves an enhancement ratio between a simulated low-resolution panchromatic (Pan) image and an original high-resolution Pan image by weighted summation of reduced HS bands.
- Robust aircraft segmentation from very high-resolution images based on bottom-up and top-down cue integration6
A segmentation method that combines bottom-up and top-down cues that can provide significant improvements on the segmentation accuracy is proposed.
- 6
- 6
- 6
- 6
- Morphological pattern-spectra-based Tau-opening optimization6
This paper investigates optimal restoration for the union noise model based upon the individual granulometric pattern spectra of the image and noise based upon a class of random-grain image and Noise models.
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- A rate-control algorithm using inter-layer information for H.264/SVC for low-delay applications5
A coarse-to-fine initial quantization parameter (QP) selection method is proposed to refine the initial QP of the EL according to the channel condition and the video sequence characteristics, and experimental results demonstrate that the proposed scheme can obtain high and smooth PSNR, and the output bit-rate is close to the target bit- rate.
- 5
- Fast Motion Detection Based on Accumulative Optical Flow and Double Background Model5
A high precise algorithm for optical flow computation with analytic wavelet and M-estimator to solve the optical flow restricted equations and introduces the extended accumulative optical flow and also provides its computational strategies, then obtains a robust motion detection algorithm.
- 4
- 4
- 4
- 4
- Saliency based multiple object cosegmentation by ensemble MIML learning4
A novel unsupervised MFC framework, which is composed of three components: un supervised label generation, saliency based pseudo-annotation and cosegmentation by MIML learning, which combines the high-level and low-level feature to represent the proposal objects, and adopts a novel SPAP clustering scheme to obtain more accurate consistent information of common objects.
- Deep Learning Based Fluid Segmentation in Retinal Optical Coherence Tomography Images4
A new deep learning based segmentation method is proposed that can adapt to the OCT scans acquired by various imaging scanning devices, and this method is more effective than other methods.
- PicMarker: Data-Driven Image Categorization Based on Iterative Clustering4
A data-driven classification method for personalized image classification tasks which can categorize images group by group and an online multi-view metric learning algorithm to learn the similarity metrics in accordance with user’s criterion is proposed.
- Research On Key Technologies Of Resource Management In Cloud Simulation Platform4
This paper proposed the guideline and architecture of the cloud simulation platform, and the resource selection technology to support the efficient allocation for types of the resources and introduces an application in the collaborative design and simulation for the multi-disciplinary virtual prototype.
- An error resilient video coding and transmission solution over error-prone channels4
A reliable error resilient video coding and video transmission framework named Z-FMO is proposed and an adaptive video transmission control algorithm JCBAF is implemented, which performs well both in R-D performance and subjective quality.
- 3
- 3
- 3
- LE-YOLO: a lightweight and efficient cucumber pruning target detection and recognition method based on improved YOLOv5s3
The LW-YOLO algorithm based on improved YOLOv5s-7.0 is proposed, which is better than the baseline model YOLOv5s-7.0 but also better than the latest models YOLOv6, YOLOv7, YOLOv7-tiny, and YOLOv8, which meet the detection and recognition requirements of cucumber pruning robots and can be deployed.
- Decoupling Degradation and Content Processing for Adverse Weather Image Restoration3
This work introduces a novel adverse weather image restoration method, called DDCNet, which decouples the degradation removal and content reconstruction process at the feature level based on their channel statistics, and exploits the unique advantages of the Fourier transform in both processes.
- 3
- Real-time object tracking via self-adaptive appearance modeling3
A self-adaptive appearance model pool based on multi-sample is constructed to improve the robustness of the object appearance models and multi-feature response fusion strategy is proposed, which can effectively improve the expression ability of the appearance model.
- 3
- Comprehensive features with randomized decision forests for hand segmentation from color images in uncontrolled indoor scenarios3
It was observed from the research that a combination of features extracting color, texture, neighborhood histogram and neighborhood probability information outperforms existing methods for hand segmentation in restricted as well as unrestricted indoor environments using just a small training dataset.
- 3
- 3
- Locality-constrained nonnegative robust shape interaction subspace clustering and its applications3
A locality-constrained nonnegative robust shape interaction (LNRSI) subspace clustering method, which guarantees the locality and the low-rank property of the optimal affinity graph and outperforms several state-of-the-art algorithms.
- 3
- An Approach to Optimized Resource Allocation for Cloud Simulation Platform3
This paper proposes a resource allocation algorithm that can maximize the overall profit of infrastructure service providers when SLA guarantees are satisfied or violated in a dynamic resource sharing cloud simulation platform and demonstrates the feasibility of the algorithm and allow a cost-effective usage of resources in cloud Simulation platform.
- 3
- 3
- Semi-Supervised Discriminant Analysis via Spectral Transduction3
A novel method called Spectral Transduction Semi-Supervised Discriminant Analysis (STSDA), which can alleviate such problem by utilizing both labeled and unlabeled data by utilizing both labeled and unlabeled data.
- The regularity analysis of multivariate refinable functions from generalized Bernstein bases and application in remote sensing image compression3
The applicability of the transforms presented in this paper to remote sensing image compression at high compression rates are explored, and the results of the experiments show that the performance is outstanding.
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- Research on High-Performance Modeling and Simulation for Complex Systems2
This chapter examines challenges of high-performance modeling and simulation for complex systems and presents solution strategies developed by the research team.
- 2
- Independent Thresholds on Multi-scale Gradient Images2
This analysis shows that proportional scale summing successfully improves edge detection rate by more than 50% and is proposed as a new multi-scale edge detection algorithm based on proportional scalesumming.
- 2
- 2
- 2
- 2
- 2
- 2
- Research on High Level Modeling Method for Complex Simulation System2
System high level modeling method is provided for describing the system’s structure and behavior which helps the system and model information to be translated among different development stages, tools and persons.
- 1
- 1
- 1
- A Pluggable Common Sense-Enhanced Framework for Knowledge Graph Completion1
This work proposes a pluggable common sense-enhanced KGC framework that incorporates both fact and common sense for KGC, and introduces common sense-guided negative sampling and a coarse-to-fine inference approach for KGs with rich entity concepts.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Finding Images by Dialoguing with Image1
This paper proposes an image retrieval oriented scene graph generation model that takes an image and a text describing the image as inputs and provides information for a newly introduced attributes head to better predict the attributes and helps constructing an adjacency matrix at the same time.
- Data Model Analysis and Integration Technology Based on Electric Power1
The experimental results show that valuable data attribute values can be obtained by using the data model analysis and the integration technology.
- A Human-Machine Collaborative Detection Model for Identifying Web Attacks1
This paper adopts a collaborative detection model, based on machine learning augmented with human interaction to detect web attacks, and designs an selection mechanism which could recommend most suspicious anomaly behaviors for humans to correct the false decision of machines.
- Automatic makeup based on generative adversarial nets1
An automatic makeup generative adversarial network is proposed, a novel approach that can perform makeup attribute editing for multiple makeup styles and is an end-to-end model with several specially designed skip layers in generator network and allows simultaneous training of multiple datasets in addition to makeup datasets.
- Saliency Detection Optimization via Modified Secondary Manifold Ranking and Blurring Depression1
An unsupervised saliency optimization method mainly via modified secondary manifold ranking and blurring depression (SMBD) that is able to promote the performances of state-of-the-art saliency detection algorithms significantly.
- Revenue-Aware Request Admission Control in Distributed Simulation Data Centers1
Experiments based on realistic requests in Google data centers demonstrate the proposed RARAC can greatly improve the revenue of SDCs provider.
- 1
- 1
- Advances in Intelligent Visual Surveillance Technology1
The processing of intelligent video surveillance based on human motion was divided into four categories: target detection, target tracking, object classification and recognition, and behavior analysis.
- Framework of A Hybrid Image Compression Algorithm for View-dependent Visualization1
A framework of a hybrid method designed to decompress image blocks of tile-pyramid used in virtual environment especially for flight simulation that combines two compression algorithms, namely DCT-based compression and WT-based one is presented.
- 1
- 1
- Vehicle classification in pan-tilt-zoom videos via sparse learning1
A simple but effective camera-invariant feature to describe the intrinsic shape patterns of vehicles can be directly extracted from two-dimensional images, eliminating the need for complicated three-dimensional template fitting used in existing vehicle classification systems.
- Measuring the Utility of Mobile Phone Video Service Based on A Service Value Network Model1
The paper presents a design for a mobile phone video service value network (MPVSVN) in China and describes a systematic approach to calculate the value and utility of services on the SVN.
- 1
- 1
- 1
- 1
- Research and Application on Virtual Prototying Engineering Grid1
The research background of Virtual Prototyping Engineering Grid (VPEG), and the connotation of VPEG are introduced in detail, including the prototyping system and some solved key technologies, some typical application demonstration systems ofVPEG.
- 1
- Eigenvector Projection Transformation and Dimension Size Reduction in Remote Sensing Data Processing1
- Micro-expression Recognition Based on Dual-branch Feature Extraction and Fusion–
This paper proposes a dual-branch micro-expression feature extraction network integrated with parallel attention that achieves 74.67% accuracy, outperforming LBP-TOP, MSMMT, and other comparative methods.
- Intelligent surface defect detection for steel manufacturing: a comparative study of YOLOv8 and Faster R-CNN in autonomous inspection systems–
YOLOv8 is more suitable for high-speed steel production lines in terms of overall accuracy, small defect detection, and real-time performance, and Faster R-CNN performs better in detecting large defects under high IoU thresholds (0.5), offering a differentiated choice for various industrial needs.
- –
- Find the stem along grape: a grape and stem segmentation method based on region images–
The “Find the Stem Along Grape” method, which locates the stem and enhances the accuracy of stem segmentation in the complex background of vineyards and contributes to intelligent grape harvesting.
- –
- Spatial-frequency attention and fusion network for few-shot fine-grained image classification–
A Spatial-frequency attention and fusion network (SFAFNet) for few-shot fine-grained image classification that not only extracts features in the spatial domain but also converts the image to the frequency domain to filter out noise interference.
- –
- –
- EPAN: Robust Pedestrian Re-Identification via Enhanced Alignment Network for IoT Surveillance–
This study introduces the Enhanced Pedestrian Alignment Network (EPAN), tailored for robust ReID across diverse IoT surveillance conditions, and demonstrates EPAN's strong feature extraction capabilities, achieving outstanding performance on the Inspection-Personnel dataset.
- –
- –
- –
- AaDR-PointCloud: An integrated point cloud processing network using attention and deep residual–
A model AaDR-PointCloud is designed that combines self-attention blocks and deep residual point blocks and operates iteratively to extract point cloud information and achieves higher accuracy on these benchmarks.
- Research on point cloud segmentation method of mutually adhered cherry tomatoes based on three-dimensional data–
An improved watershed segmentation method based on point cloud spatial density cluster analysis is proposed, which improves the three-dimensional segmentation performance of stuck cherry tomatoes.
- –
- –
- –
- –
- –
- Surface defect detection of the power machinery based on deep learning–
This paper investigates the performance of various convolutional neural network models in the task of detecting surface defects, identifies relatively excellent detection models based on the confusion matrix, and summarizes the more appropriate detection scenarios for different models.
- –
- –
- –
- –
- An entropy based unsupervised visual vocabulary optimization for visual loop closure detection–
An entropy ranking based scheme to improve the spectral clustering algorithm by leveraging eigenvector selection and feature embedding simultaneously and results demonstrate that the proposed BoW based feature representation has strong representation ability in measuring the similarity of loop closure events and differentiating new visited places.
- –
- –
- –
- –
- –
- Naive Gabor Networks.–
This paper introduces naive Gabor Networks or Gabor-Nets, which, for the first time in the literature, design and learn convolutional kernels strictly in the form of Gabor filters, aiming to reduce the number of parameters and constrain the solution space for Convolutional neural networks (CNNs).
- Learning salient seeds refer to the manifold ranking and background-prior strategy–
A brand-new object detection method considering both local and global features, based on the concept of manifold ranking, is designed, which is able to lower down the detection error rate in case that the object located near the boundary.
- –
- Research of Traffic Flow Based on SVM Method–
A new method combined Enhanced binarization method with SVM (Support Vector Machine) Machine learning mode is proposed to improve the accuracy of the recognition scheme of the headlights.
- –
- –
- Precursor Gas Sensor Detection and Recognition Based On Metrology Method–
This paper designed and developed a chromatographic separation and sensor based on the combination of gas detection instruments, to multi gas detection instrument, to solve the traditional single common precursor gas detection.
- 2D Visual odometry based on Probability Data Association Filter–
This paper presents a novel approach based on probability data association (PDA) filtering for estimating a vehicle's trajectory in complex urban environments that provides dynamic estimation of ego-motion vector in the presence of falsely feature pairs.
- Research on Vehicle Detection and Shadow Elimination Based on Static Image Sequence–
A new approach for vehicle detection and shadow elimination based on average background modeling is proposed, which uses average background model and background subtraction to locate vehicle roughly, eliminates shadow of the vehicle using canny edge detection with dynamic histogram threshold determined by the histogram of the image.
- The Improvement and Implementation of Examinee Identification Algorithm Based on the Video Surveillance–
This paper proposes the examinee identification algorithm based on the video surveillance that is simple, has a good real-time performance and robustness, and can detect whether examinee in his place by using the comprehensive method which is background subtraction and the normalized matching algorithm.
- A Novel Medical Image Registration Based on Contour Features and MultiScale Information–
A novel hierarchical registration method of the medical images based on multiscale information and contour line is proposed and can achieve more accurate registration results, and can reduce the computational time greatly.
- –
- –
- –
- A Design of Mobile Payments Business Model Based on Value Network–
The design of business model of mobile payments was divided into four phases: value discovery, value creation, value management, and value realization, and every phase was designed in detail.
- A Low-Distortion Single Fingerprint Acquisition Optical System–
A method of correction for the trapezoid distortion of the single fingerprint acquisition system is proposed, which combines nearly telecentric optical path and the inclined plane compensation.
- Research on co-simulation task scheduling in Cloud Simulation Platform–
In order to address the problem of single collaborative simulation task scheduling, considering the characteristics of simulation resource encapsulated with virtualization technology, the author first proposed the unified model describing the co-simulation system for co-Simulation task, which is the basis of task scheduling.
- –
- –
- –
- –
- –
- Research on Simulation Application Mode and Run-Time Framework in Simulation Grid–
The simulation application mode in SimGrid environment is introduced, which puts emphasis on simpleness for user to implement simulation application, and requires SimGrid supplying ability on self-organization and adaptiveness in implement of simulation system.
- –
- –
- The research on image enhancement of low-quality fingerprint image–
A fingerprint enhancement algorithm based on Fourier filtering is proposed that is fast and has excellent enhancing performance for low quality fingerprints, especially for noisy or low quality images.
- Wavelet image compression based on bit-plane context and its application to remote sensing images–
Experimental results show that the scheme is among the best image coders for natural images and remote sensing images.
- A fingerprint recognition method based on Fourier filtering enhancement and minutia matching–
A minutia matching algorithm based on Fourier filtering that introduced ridge information into the minutian matching process in a simple but effective way and solved the problem of the matching of vector pairs with low computational cost.
- –
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 2026-10-14. 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.