Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks217 from public data
- WENETSPEECH: A 10000+ Hours Multi-Domain Mandarin Corpus for Speech Recognition394
WenetSpeech is the current largest open-source Mandarin speech corpus with transcriptions, which benefits research on production-level speech recognition, and a novel end-to-end label error detection approach is proposed.
- 381
- M2Met: The Icassp 2022 Multi-Channel Multi-Party Meeting Transcription Challenge221
The AliMeeting corpus, which consists of 120 hours of recorded Mandarin meeting data, is made available and the ICASSP 2022 Multi-channel Multi-party Meeting Transcription Challenge (M2MeT) is launched with two tracks, namely speaker diarization and multi-speaker ASR, aiming to provide a common testbed for meeting rich transcription and promote reproducible research in this field.
- AISHELL-4: An Open Source Dataset for Speech Enhancement, Separation, Recognition and Speaker Diarization in Conference Scenario187
AISHELL-4, a sizable real-recorded Mandarin speech dataset collected by 8-channel circular microphone array for speech processing in conference scenario, is presented, and is the only Mandarin dataset for conversation speech, providing additional value for data diversity in speech community.
- 168
- Comprehensive Analysis of Fibroblast Activation Protein Expression in Interstitial Lung Diseases152
It was revealed, for the first time, FAP expression was upregulated significantly in the early phase of lung fibroblast activation event in response to a low dose of profibrotic cytokine, and the total standard uptake value of FAP inhibitor (FAPI) PET (SUVtotal) was significantly related to lung function decline in patients with ILD.
- An MRI-based atlas and database of the developing mouse brain147
DTI based atlases of ex vivo C57BL/6 mouse brains at several developmental stages are presented, enhancing the ability to examine the neuroanatomy in normal or genetically engineered mouse strains and mouse models of neurological diseases.
- Exploring Image Enhancement for Salient Object Detection in Low Light Images117
The proposed model directly embeds the physical lighting model into the deep neural network to describe the degradation of low light images, in which the environment light is treated as a point-wise variate and changes with local content.
- Resource atlases for multi-atlas brain segmentations with multiple ontology levels based on T1-weighted MRI112
The details of the atlas resources are described and the improved accuracy achievable with a dynamic age-matching approach, in which atlases that most closely match the subject’s age are dynamically selected are demonstrated.
- 99
- Digital 3D Brain MRI Arterial Territories Atlas95
This study provides the first publicly-available, digital, 3D deformable atlas of arterial brain territories, which may serve as a valuable resource for large-scale, reproducible processing and analysis of brain MRIs of patients with stroke and other conditions.
- MuseCoco: Generating Symbolic Music from Text86
MuseCoco stands for Music Composition Copilot that empowers musicians to generate music directly from given text descriptions, offering a significant improvement in efficiency compared to creating music entirely from scratch.
- BiRA-Net: Bilinear Attention Net for Diabetic Retinopathy Grading86
A new deep learning architecture, called BiRA-Net, is proposed, which combines the attention model for feature extraction and bilinear model for fine-grained classification, and a new loss function, called grading loss, which leads to improved training convergence of the proposed approach.
- 79
- 77
- 75
- 57
- A survey of CAPTCHA technologies to distinguish between human and computer53
This review discusses the CAPTCHA mechanism from usability and security aspects, therefore attacking (anti-classification) and defending ( classification) technologies towards currentCAPTCHA are both reviewed.
- Robust Automatic Focus Algorithm for Low Contrast Images Using a New Contrast Measure53
First, a noise reduction preprocessing is introduced to make the algorithm robust to both additive noise and multiplicative noise, and a new contrast measure is presented to bring in local false peaks, ensuring the presence of a well defined focused peak.
- Large-scale gesture recognition with a fusion of RGB-D data based on optical flow and the C3D model46
An effective 3D Convolutional Neural Network based method for large-scale gesture recognition using RGB-D video data that achieves 54.50% accuracy on the validation subset and 60.93% on the testing subset of the Chalearn LAP IsoGD dataset, both of which outperform the proposed method's results.
- Summary on the ICASSP 2022 Multi-Channel Multi-Party Meeting Transcription Grand Challenge44
The ICASSP 2022 Multi-channel Multi-party Meeting Transcription Grand Challenge (M2MeT) focuses on one of the most valuable and the most challenging scenarios of speech technologies, and releases 120 hours of real-recorded Mandarin meeting speech data with manual annotation.
- Rank-in-Rank Loss for Person Re-identification38
Extensive experimental results indicate that the proposed Differentiable Retrieval-Sort Loss (DRSL) can improve the performance of the state-of-the-art re-ID models, thus demonstrating its effectiveness and superiority in the re- ID task.
- A robust detail preserving anisotropic diffusion for speckle reduction in ultrasound images38
The proposed robust detail preserving anisotropic diffusion filter (RDPAD) for speckle reduction can preserve the important structure information of the original image while reducing speckles and is also less sensitive to the number of iterations.
- 36
- The Multi-Speaker Multi-Style Voice Cloning Challenge 202135
The challenge to adapt an average TTS model to the stylistic target voice with limited data from target speaker, evaluated by speaker identity and style similarity is formulated.
- FMRNet: Image Deraining via Frequency Mutual Revision34
This work proposes to investigate the potential relationships among rain-free and residue components at the frequency domain, forming a frequency mutual revision network (FMRNet) for image deraining, and explores the mutual representation of rain residue and background components at frequency domain.
- 34
- 31
- 29
- 29
- Learning a Unified Blind Image Quality Metric via On-Line and Off-Line Big Training Instances29
Through extensive experiments, it has been validated that the proposed blind IQM is capable of simultaneously inferring the quality of NS and SC images, and it has attained superior performance as compared with popular and state-of-the-art IQMs on the subjective NS andSC image quality databases.
- 29
- 27
- RETRACTED ARTICLE: Salient object detection using a covariance-based CNN model in low-contrast images27
This paper proposes a novel deep neural network framework embedded with covariance descriptor for salient object detection in low-contrast images via a 7-layer convolutional neural network (CNN).
- Cross-Modality Retrieval by Joint Correlation Learning26
This article proposes a cross-modal learning model with joint correlative calculation learning that achieves comparable performance to the state-of-the-art on three benchmarks, i.e., Flickr8k, Flickr30k, and MS-COCO.
- 26
- 24
- 24
- 24
- A cross-media distance metric learning framework based on multi-view correlation mining and matching24
This paper proposes a novel cross-media distance metric learning framework based on sparse feature selection and multi-view matching which harmonizes different correlations with an iteration process and builds Cross-media Semantic Space for cross- media distance measure.
- 23
- Longitudinal imaging highlights preferential basal ganglia circuit atrophy in Huntington’s disease22
The results support (but certainly do not prove) the hypothesis of circuit-based spread of pathology in Huntington’s disease, possibly due to spread of mutant Htt protein, though other connection-based mechanisms are possible.
- Salient object detection from low contrast images based on local contrast enhancing and non-local feature learning22
A salient object detection model based on local contrast enhancing and non-local feature learning that yields competitive performance compared to existing state-of-the-art models is presented.
- Classroom Attention Analysis Based on Multiple Euler Angles Constraint and Head Pose Estimation22
A scoring module for converting the head Euler angle and attention in the classroom by introducing the mutual constraint of multiple Euler angles with the head spatial information, aiming to make attention model less susceptible to the difference of head information.
- 22
- 22
- PointNet on FPGA for Real-Time LiDAR Point Cloud Processing20
PointNet, the state-of-art deep learning algorithm for point cloud processing, is successfully implemented on the proposed FPGA platform and meets the real-time requirement for most of the existing LiDAR sensors.
- 19
- 19
- 19
- VAGeo: View-specific Attention for Cross-View Object Geo-Localization18
A novel View-specific Attention Geo-localization method (VAGeo) for accurate CVOGL, which contains two key modules: view-specific positional encoding (VSPE) module and channel-spatial hybrid attention (CSHA) module.
- 18
- Unsupervised Video Person Re-Identification via Noise and Hard Frame Aware Clustering18
Experiments conducted on two video-based datasets demonstrate the effectiveness of the proposed NHAC under the unsupervised re-ID setting.
- ESOD-YOLO: an enhanced efficient small object detection framework for aerial images17
ESOD-YOLO, an efficient detector based on YOLOv8 that optimally balances accuracy and efficiency is proposed in this work, and results demonstrate the efficiency of the proposed innovative approach.
- 17
- Consistency-Constancy Bi-Knowledge Learning for Pedestrian Detection in Night Surveillance17
A consistency-constancy bi-knowledge learning (CCBL) for pedestrian detection in night surveillance is proposed, which is able to simultaneously achieve the night pedestrian detection's useful knowledge, coming from day and night surveillance.
- Dust particle detection in surveillance video using salient visual descriptors17
The proposed approach is able to achieve superior dust particle detection performance to that of conventional approaches, as evaluated in real-world video surveillance scenarios.
- Hierarchical salient object detection model using contrast-based saliency and color spatial distribution17
A hierarchical salient object detection model is presented that can accurately detect the salient object which conform with the human visual principle and the experimental results from the MSRA database validate the effectiveness of the model.
- Low-light image enhancement with joint illumination and noise data distribution transformation16
This work proposes a novel low-light image enhancement method based on simultaneous adjustment on illumination and noise using unpaired data that can reach the network capability that trained with rich paired data and achieved satisfactory results.
- Saliency Detection Based on the Combination of High-Level Knowledge and Low-Level Cues in Foggy Images16
A novel saliency detection method for human tracking in the wild that can remove the influence of fog efficiently, and the performance is better than 16 state-of-the-art saliency models.
- Discrete stationary wavelet transform based saliency information fusion from frequency and spatial domain in low contrast images16
This paper proposes a salient object detection model utilizing efficient features both from frequency domain and spatial domain in low contrast images and demonstrates the effectiveness of the proposed saliency detection model over other ten state-of-the-art saliency models.
- 15
- Multi-level feature fusion model-based real-time person re-identification for forensics15
Experimental results indicate MFFM can achieve the best performance compared to the state-of-the-art models on the Market1501, CUHK03, and VIPeR datasets.
- Mix-Modality Person Re-Identification: A New and Practical Paradigm13
A Mix-Modality Person Re-Identification (MM-ReID) task is proposed, the influence of modality mixing ratio on performance is explored, and mix-modality test sets for existing datasets are constructed according to the new mix-modality testing paradigm.
- 13
- 13
- End-to-end video subtitle recognition via a deep Residual Neural Network13
An end-to-end pipeline for recognizing subtitles of video, where Connectionist Text Proposal Network is utilized for video subtitle detection, while Residual Network, Gated Recurrent Unit, and Connectionist Temporal Classification are used to recognize Chinese and English subtitles in video images.
- 13
- Extended Non-local Feature for Visual Saliency Detection in Low Contrast Images12
The key idea of the proposed model is to hierarchically introduce non-local module with local contrast processing blocks, aiming to provide robust representation of saliency information towards low contrast images with low signal-to-noise ratio property.
- 12
- Screen-rendered text images recognition using a deep residual network based segmentation-free method12
- Lightweight CNN-Based Low-Light-Image Enhancement System on FPGA Platform11
A lightweight convolutional neural network for low light image enhancement task is proposed with less than 1 M parameter size and solving the problem of hyper parameter setting in enhancement models.
- 11
- Extremely efficient PM2.5 estimator based on analysis of saliency and statistics11
The authors extract two features of the entropy and intensity values of non-salient image saturation map into the approximation of PM2.5 concentration and show that the proposed IPPS is superior in accuracy and efficiency.
- 10
- Efficient virtual data search for annotation‐free vehicle reidentification10
This paper presents a novel data sampling strategy from both semantic and feature levels to perform an effective data search and is among the first attempts to search effective virtual data to perform annotation‐free vehicle re‐ID.
- Multiple kernel visual-auditory representation learning for retrieval10
This paper proposes a multiple kernel visual-auditory representation learning approach, which learns cross- media correlations from visual and auditory feature spaces with multiple kernel strategies, and gives cross-media distance measure for image-audio retrieval in the mutual subspace of co-occurrence.
- 10
- 10
- Informative Classes Matter: Towards Unsupervised Domain Adaptive Nighttime Semantic Segmentation9
An Online Informative Class Sampling (OICS) strategy to adaptively mine informative classes from the target nighttime domain according to the corresponding spectrogram mean and the class frequency via the authors' Informative Mixture of Experts framework, which outperforms state-of-the-art UDA-NSS methods by large margins on three widely-used benchmarks.
- Chinese Characters Recognition from Screen-Rendered Images Using Inception Deep Learning Architecture9
The proposed Chinese characters recognition model using inception module based convolutional neural networks can effectively segment Chinese characters from screen-rendered images, and significantly reduce the training time.
- Blind 3D Video Stabilization with Spatio-Temporally Varying Motion Blur8
The Cross-frame-temporal Module framework is proposed to address blind motion blur induced by various conditions, which utilizes cross-frame temporal features to estimate depth maps and camera motion and outperforms state-of-the-art methods.
- ICMC-ASR: The ICASSP 2024 In-Car Multi-Channel Automatic Speech Recognition Challenge8
The ICASSP 2024 In-Car Multi-Channel Automatic Speech Recognition Challenge is launched and first-place team USTC-iflytek achieves a CER of 13.16% in the ASR track and a cpCER of 21.48% in the ASDR track, showing an absolute improvement of 13.08% and 51.4% compared to the challenge baseline, respectively.
- 8
- A Novel Infrared Small Target Detection Algorithm Based on Deep Learning8
A novel deep learning-based infrared small target detection algorithm called Single shot Detector for Small Target (SSD-ST) is proposed by redesigning the network architecture of Single Shot Detector (SSS) by dropping the low-resolution layers and enhancing the high-resolution layer in SSD to make it more suitable for the detection task of infrared small targets.
- Finding autofocus region in low contrast surveillance images using CNN-based saliency algorithm8
A covariance based deep learning framework is proposed to evaluate the saliency of low contrast surveillance image and outperforms existing ten state-of-the-art saliency models on three public datasets and a nighttime image dataset.
- 8
- Solid oxide fuel cell anode image segmentation based on a novel quantum-inspired fuzzy clustering8
A novel image segmentation method on YSZ/Ni anode Optical Microscopic images provides firm foundation to investigate the microstructural evolution and its related properties and improves accuracy and effectiveness of three-phase identification on the micro-investigation.
- Parts-based face super-resolution via non-negative matrix factorization8
The proposed CCNMF algorithm can generate global face more similar to the ground truth face by learning a parts-based local representation of facial images; while the HCNMF can learn the relation between high-resolution residue and low- resolution residue to better preserve high frequency details.
- Robust passive autofocus system for mobile phone camera applications8
A passive autofocus system with low computational complexity that uses a novel modified Entropy based contrast measurement to determine degree of image sharpness, which can better reflect the information about image discontinuities.
- ICLR: Inter-Chrominance and Luminance Interaction for Natural Color Restoration in Low-Light Image Enhancement7
Experimental results show that the proposed ICLR framework outperforms state-of-the-art methods in improving contrast while restoring details and textures for images captured in low-light conditions.
- 7
- 7
- 7
- Road Pothole Recognition and Size Measurement Based on the Fusion of Camera and LiDAR6
This work proposes a fusion-based approach that integrates images and LiDAR point clouds for measuring pothole dimensions, achieving a 27.4% improvement in accuracy over single LiDAR methods and reducing the average processing time by 88.2%.
- Glare countering and exploiting via dual stream network for nighttime vehicle detection6
A novel dual-branch network for processing only RGB images with glare in nighttime vehicle detection, which uses only RGB images, does not introduce additional information, and achieves comparable performance to methods based on infrared images.
- A comprehensive survey of visible infrared person re-identification from an application perspective6
A novel perspective is offered by classifying VI-ReID according to application scenarios by categorizing diverse methodologies under the proposed classification and assessing their performance metrics across common datasets.
- Identification of prognostic and driver gene mutations in acute myeloid leukemia by a bioinformatics analysis6
The systemic analysis of the gene mutation in patients with AML was performed and representative and driver mutations between the prognostic group were identified, helping physicians better predict patient prognosis and thus develop better treatment plans.
- Nonnegative cross-media recoding of visual-auditory content for social media analysis6
This paper proposes a novel nonnegative cross-media recoding approach, which learns co-occurrences of cross- media feature spaces by explicitly learning a common subset of basis vectors and imposes the nonnegativity constraint on the decomposed matrices so that the basis vectors represent important and locally meaningful features of the cross- Media data.
- Mass classification in mammogram with semi-supervised relief based feature selection6
The experimental result shows that by incorporating information embedded in unlabeled data, SSLFE can improve the performance compared to the method without feature selection and traditional Relief method.
- 5
- SEANet: Semantic Enhancement and Amplification for Underwater Object Detection in Complex Visual Scenarios5
The proposed Semantic Enhancement and Amplification Network (SEANet), a framework designed to enhance underwater object detection in complex visual scenarios, achieves the highest AP (Average Precision) of 67.0% on the RUOD dataset, 53.0% on the URPC2021 dataset, and 71.5% on the DUO dataset.
- An Adaptive ICP LiDAR Odometry Based on Reliable Initial Pose5
This article proposes an adaptive ICP-based LiDAR odometry method that relies on a reliable initial pose and outperforms existing approaches and significantly enhances the accuracy of LiDAR odometry.
- The Parables of the Mustard Seed and the Yeast: Extremely Low-Budget, High-Performance Nighttime Semantic Segmentation5
A new task focusing on exploring the potential of training strategy and framework design with limited annotation to achieve high-performance NSS, and an innovative Fragmentation Synergy Active Domain Adaptation framework (FS-ADA), which progressively broadcasts the limited annotation to the unlabeled regions, achieving high performance with a minimal annotation budget.
- Towards Causality Inference for Very Important Person Localization5
A causal inference framework is established that unpacks the causes of previous methods and derives a new principled solution for VIPLoc, which treats the scene as confounding factor, allowing the ever-elusive confounding effects to be eliminated and the essential determinants to be uncovered.
- The ISCSLP 2022 Intelligent Cockpit Speech Recognition Challenge (ICSRC): Dataset, Tracks, Baseline and Results5
This paper summarizes the outcomes from the ISCSLP 2022 Intelligent Cockpit Speech Recognition Challenge (ICSRC) and introduces the associated dataset collected from a new-energy vehicle covering a variety of cockpit acoustic conditions and linguistic contents.
- Active learning with sampling by joint global-local uncertainty for salient object detection5
A novel active sampling strategy for salient object detection that can be reduced by 48.3% with competitive performance compared with the state-of-the-art SOD model.
- Visible-Infrared Cross-Modal Person Re-identification based on Positive Feedback5
This paper presents the idea of cross-modality visible-infrared person re-identification combined with human positive feedback, and shows that the proposed method achieved superior performance compared to the current state-of-the-art methods.
- Optimal Feature Selection for Saliency Seed Propagation in Low Contrast Images5
An optimal feature selection based saliency seed propagation method is presented to detect salient objects in low contrast images and outperforms 15 state-of-the-art methods in terms of efficiency and accuracy.
- Estimation of Local and Global Superpixel Covariance for Salient Object Detection in Low Contrast Images5
This paper proposes to estimate the saliency of low contrast images via covariance features, where the input image is firstly decomposed into superpixel regions to estimate their covariances, and the local and global image saliency can be calculated using the covariance Features respectively.
- Mass Classification with Level Set Segmentation and Shape Analysis for Breast Cancer Diagnosis Using Mammography5
The experimental results show that Fourier descriptor of normalized accumulative angle is an effective feature for the classification of masses in mammogram.
- 4
- Positive Style Accumulation: A Style Screening and Continuous Utilization Framework for Federated DG-ReID4
A style memory recognition loss is proposed to fully leverage the positive styles memorized by the Generalization Gain-guided Dynamic Style Memory, and a Collaborative Style Training strategy is proposed to make full use of positive styles.
- 4
- Unsupervised Low-Light Image Enhancement Based on Curve Estimation and Illumination Perception4
This paper proposes an end-to-end low-light image-enhancing method, which does not require paired datasets with ground truth and uses a lightweight deep neural network that estimates curve parameters and uses a set of no-reference loss functions to assess the quality of the enhanced image.
- 4
- Translating clinical notes into quantitative measures—a real-world observation on the response to cholinesterase inhibitors or selective serotonin reuptake inhibitors prescribed to outpatients with dementia using electronic medical records4
The results were remarkably similar to what was reported in controlled clinical trials of Alzheimer’s disease and its related neuropsychiatric symptoms.
- 4
- A Novel Method of Detecting Chinese Rendered Text on Tilted Screen of Mobile Devices4
A Chinese-style scene text detection method for tilting screens capturing by mobile devices using the Gaussian removal method to delete the narrow word width and designed a more accurate network to recognize characters is proposed.
- Cross-media retrieval based on semi-supervised regularization and correlation learning4
This work proposes a novel feature learning algorithm for cross-media retrieval, called semi-supervised regularization and correlation learning (SSRCL), which is capable of modeling multiple types of media simultaneously, and considers both semantic category similarity and dissimilarity simultaneously.
- 4
- An automatic focus algorithm for still and video camera applications using a new contrast measure4
An auto-focus algorithm based on the new defined sharpness function is developed that is more robust to noise, jitter of the camera and has better performance.
- 4
- 3
- MambaOVSR: Multiscale Fusion with Global Motion Modeling for Chinese Opera Video Super-Resolution3
This work pioneer a large-scale Chinese Opera Video Clip dataset and proposes the Mamba-based multiscale fusion network for space-time Opera Video Super-Resolution (MambaOVSR), which significantly outperforms the SOTA STVSR method in terms of PSNR.
- Select Informative Samples for Night-Time Vehicle Detection Benchmark in Urban Scenes3
A vehicle detection framework via an active auxiliary mechanism (AAM) to reduce the annotation workload and a computer-assisted detection module embedded in the AAM framework to help human annotators to rapidly and accurately label the selected data.
- 3
- 3
- Dual-focus: person search from Coarse-Grained Focus to Fine-Grained Focus3
Inspired by the observation that the police first narrowed the scope to a small group of people similar to the suspect, and then located the target, this work proposes a Dual-Focus framework consisting of Coarse-Grained Focus (search for similar people) and Fine-Graining Focus (find the target).
- 3
- Target group distribution pattern analysis with bagged convolutional neural networks for UAV distribution pattern identification3
Experimental results indicate that the method is robust to target missing and location variance and scalable with dataset size, and has outperformed the benchmark machine learning methods significantly in pattern identification accuracy.
- 3
- A spatial-frequency-temporal domain based saliency model for low contrast video sequences3
A dynamic multimodal fusion based visual saliency detection model towards low contrast videos, which combines saliency information from spatial, frequency, and temporal domains is proposed.
- 3
- Inferring Visual Perceptual Object by Adaptive Fusion of Image Salient Features3
A novel data driven approach is proposed to adaptively select proper features for different kinds of images using low-level features containing the most distinguishable salient information per image based on the adaptive weight selection scheme.
- Color Image Segmentation Algorithm of Rapid Level Sets Based on HSV Color Space3
The color image segmentation of rapid level sets based on HSV color space based on CV mode is put forward and shows it is more effective in segmentation than the traditional segmentation method of level sets.
- 3
- 2
- Foundation Model Insights and a Multi-Model Approach for Superior Fine-Grained One-shot Subset Selection2
Surprising insights are uncovered: FMs consistently outperform traditional IEs on fine-grained datasets, whereas their advantage diminishes on coarse-grained datasets with noisy labels.
- 2
- Style Separation and Content Recovery for Generalizable Sketch Re-identification and a New Benchmark2
- 2
- Adversarial Recovery Network for Low-Light Person Re-Identification2
An Adversarial Recovery Network (ARN) is proposed that enables a re-ID model with low-light recognition capability without the need for low-lights data, and outperforms other competitors by a wide margin.
- 2
- Deep Camera‐Aware Metric Learning for Person Reidentification2
A deep camera-aware metric learning (DCAML) model, where images on the identity-level spaces are further projected into different camera-level subspaces, which can explore the inherent relationship between identity and camera.
- Study on Detection Image Processing Method of Offshore Cage2
The adaptive threshold segmentation algorithm combined with mathematical morphology processing method is used to segment the offshore cage image, and extract the skeleton of the decomposed image to realize the detection of the integrity of the offshore net.
- An on-line path planning of assembly robots based on FOVS2
An on-line path planning method for mechanical assembly process robots based on visual field space based on field-of-view space (FOVS) is proposed and the result shows that the algorithm is effective.
- 2
- 2
- 2
- Learn a Deep Convolutional Neural Network for Image Smoke Detection2
A 10-layer convolutional neural network structure designed for smoke characteristics is proposed that has fewer parameters than the existing deep learning method, and the accuracy rate in the smoke database is optimal.
- Three phases segmentation from Ni/YSZ anode optical microscopy images using quantum-inspired mixture clustering model2
The experimental results demonstrate that the proposed mixture clustering model is able to achieve a higher segmentation accuracy with a faster convergence speed, which lays firm foundation for microstructural parameters extraction from SOFC electrodes image datasets.
- 2
- 2
- Particle Swarm Optimization Based Salient Object Detection for Low Contrast Images2
Experimental results show that the proposed salient object detection model yields better results than eleven state-of-the-art saliency models on low contrast images.
- Latent semantic factorization for multimedia representation learning2
A novel multimedia representation learning framework via latent semantic factorization (LSF), where the posterior probability under the learned classifiers is served as the latent semantic representation for different modalities.
- 2
- Salient Object Detection from Single Haze Images via Dark Channel Prior and Region Covariance Descriptor2
This paper proposed to use a computational model of visual saliency to cope with object detection from single haze images and verified the validity and efficiency of the proposed saliency computational model.
- Hierarchical Features Fusion for Salient Object Detection in Low Contrast Images2
This paper proposes a saliency detection model based on the feature extraction, in which optimal features have been learned in different contrast environment to improve the effectiveness of salient object detection.
- 2
- Mass Diagnosis in Mammography with Mutual Information Based Feature Selection and Support Vector Machine2
The experimental result shows that mutual information based feature selection is useful for the diagnosis of masses as well as Linear discriminant analysis and support vector machine are investigated for the final classification.
- 2
- Adaptive Variance Based Sharpness Computation for Low Contrast Images2
An improved variance based sharpness computation which can adapt to various degrees of noise and bring in the local false peaks generated by noise influence, and therefore produce a well defined focused peak standing for the best focused image.
- 2
- 1
- 1
- 1
- FIMD: Fusion-Inspired Modality Distillation for Enhanced MRI Segmentation in Incomplete Multi-Modal Scenarios1
The effectiveness of the FIMD framework in enhancing the robustness and accuracy of brain tumor segmentation is demonstrated and the FIMD method, devoid of specific architectural constraints, seamlessly integrates into existing multi-modal brain tumor segmentation frameworks.
- Alleviating Long-Tailed Image Classification via Dynamical Classwise Splitting1
This work proposes a novel classwise splitting (CWS) method built upon a dynamic cluster, where classwise prototypes are updated using a moving average technique, and shows that when trained with the proposed class-balanced loss, the network is able to achieve significant performance gains on long-tailed datasets.
- Semisupervised Cross-Media Retrieval by Distance-Preserving Correlation Learning and Multi-modal Manifold Regularization1
A novel distance-preserving correlation learning and multi-modal manifold regularization approach to exploit the common representation of heterogeneous data for cross-media retrieval, compared with the state-of-the-art methods.
- 1
- CBR Based Educational Method for the Postgraduate Course Image Processing and Machine Vision1
A CBR based method to construct educational system is proposed to address the problem of deepen teaching innovation and to further improve the teaching quality and effect of the image processing and machine vision postgraduate courses.
- 1
- Fuzzy Approach Based on Quantum-Inspired MRF for SOFC Anode Optical Microscope Images Segmentation1
An adaptive fuzzy approach is developed for three-phase identification of YSZ/Ni anode Optical Microscopic images and a new quantum-inspired clique potential Markov random field (MRF) function is proposed to considerate spatial information in fuzzy logic model.
- Superpixel Level Incremental Visual Saliency Detection in Low Contrast Video1
A saliency detection model to detect salient object in low contrast video, which combines spatial saliency computation with temporalSaliency computation, and incremental learning algorithm is employed in temporal domain to effectively update the background model.
- A multiscale superpixel-level salient object detection model using local-global contrast cue1
A salient object detection model is proposed by integrating local and global superpixel contrast at multiple scales by integrating the saliency of superpixel to refine the resulting saliency map.
- Mutual information optimization based dynamic log-polar image registration1
Experimental results show that the combination of MI with DLPT is an effective and precise method for image registration.
- 1
- Exploiting local and global characteristics for contrast based visual saliency detection1
A visual saliency detection algorithm based on a novel contrast measurement that extracts the spectral information of image block using the 2D discrete Fourier transform (DFT), and combines with the total variation (TV) of image block in spatial domain is presented.
- 1
- 1
- –
- Fusing LabVIEW and Machine Learning: A Project‐Based Approach for Teaching Industrial Condition Monitoring–
The module successfully demonstrates a scalable framework for incorporating AI into engineering laboratories, effectively preparing students for roles that require synthesizing physical system knowledge with intelligent algorithm deployment.
- –
- –
- Structurally Tunable High‐Thermal‐Conductivity Silicon Carbide Composites for Thermal Energy Storage–
- –
- –
- –
- Latent Watermarking: Secure Cryptographic Hashing in Flow-Matching for AI Content Authentication–
This approach addresses the limitations of prior techniques, overcoming inversion-induced distortion while enabling efficient embedding and decoding of meaningful authentication messages without pretraining or fine-tuning.
- –
- –
- Diversity-Representativeness Replay and Knowledge Alignment for Lifelong Vehicle Re-identification–
A novel Diversity-Representativeness Dual-Stage Sampling Replay (DDSR) strategy for LVReID that constructs an effective memory buffer through two stages, i.e., Cluster-Centric Class Selection and Diverse Instance Mining and introduces Maximum Mean Discrepancy loss to align the feature distribution between replay data and the new arrivals.
- –
- –
- –
- –
- –
- LIMPID: A Lightweight Image-to-Curve MaPpIng moDel for Extremely Low-Light Image Enhancement–
The image-to-curve transformation to ELLIE is introduced for the first time and a new pyramid fusion strategy based on Laplacian and Gaussian is proposed, which attempts to achieve dynamic fusion of multi-scale images via learnable fusion weight parameters.
- –
- –
- Variance Weight Distribution Network Based Noise Sample Learning for Robust Person Re-identification–
- –
- –
- –
- Method for Creating, Updating and Maintaining a Case Library of Service Business Guidance–
This article provides a method for creating, updating and maintaining a case library of service business guidance that uses genetic algorithm to filter and summarize cases that have been processed in the service business area to form a guidance case library.
- –
- A Quantum-Inspired Fuzzy Clustering for Solid Oxide Fuel Cell Anode Optical Microscope Images Segmentation–
A quantum-inspired adaptive fuzziness factor is introduced to adaptively estimate the parameters of the spatial constraint term in the fuzzy clustering based on Markov Random Filed (MRF) for YSZ/Ni anode Optical Microscopic images.
- –
- Saliency Detection Model for Low Contrast Images Based on Amplitude Spectrum Analysis and Superpixel Segmentation–
A superpixel based feature extraction method is utilized to generate saliency map via both local and global spatial information to demonstrate the effectiveness of the proposed saliency detection model over other eight state-of-the-art saliency models.
- Improved Dendritic Cell Algorithm with False Positives and False Negatives Adjustable–
Experiments show that the two kinds of DCA which can adjust false positives and false negatives can both effectively realize results controlled, comparatively the scoring DCA is more intuitive.
- Superpixel-Based Global Contrast Driven Saliency Detection in Low Contrast Images–
The proposed model perfectly extracts the salient object in low contrast visibility conditions, which has been tested on three public datasets, as well as a nighttime image dataset, and outperforms existing state-of-the-art saliency detection models.
- –
- SAR Image Filtering Based on Quantum-Inspired Estimation of Speckle Variance–
A novel filtering method is proposed for synthetic aperture radar (SAR) images by applying the dual-tree complex wavelet transform (DTCWT) with an adaptive quantum-inspired estimation of speckle noise variance and an adaptive despeckling method according to the Bayesian maximum a posteriori (MAP) theory is proposed.
- –
- –
- Image Processing System of Automatic Radiographic Testing for Fuse Assemblage–
Based on the research of digital image collection, image processing and automatic recognition, the fuse X-ray image automatic detection and recognition was realized and the fuse assemblage automatic detection was realized in the system.
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.