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
- Suggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works232 from public data
- 436
- 204
- 174
- 157
- 94
- Person re-identification with content and context re-ranking90
A novel and efficient re-ranking technque to solve the person re-identification problem in the surveillance application and can be divided into offline and online parts, where the majority of computation load is accomplished by the offline part and the online computation complexity is only proportional to the size of the gallery data set.
- One More Check: Making “Fake Background” Be Tracked Again87
This paper constructs a new one-shot tracker and reaches a new state-of-the-art MOTA and IDF1 performance and proposes a re-check network, which innovatively expands the role of ID embedding from data association to motion forecasting by effectively propagating previous tracklets to the current frame with a small overhead.
- 87
- 82
- 72
- 65
- Image classification by non-negative sparse coding, correlation constrained low-rank and sparse decomposition65
An image classification framework by leveraging the non-negative sparse coding, correlation constrained low rank and sparse matrix decomposition technique (CCLR-Sc+SPM), which achieves or outperforms the state-of-the-art results on several benchmarks.
- 63
- Equidistance constrained metric learning for person re-identification61
An algorithm for learning a Mahalanobis distance for person re-identification that outperforms the state-of-the-art methods on CUHK01, CUHK03, Market1501 and DukeMTMC-reID datasets, and achieves very competitive performance on the widely used VIPeR dataset.
- 56
- Ranking Optimization for Person Re-identification via Similarity and Dissimilarity56
This paper proposes to use both similarity and dissimilarity cues in a ranking optimization framework for person re-identification, based on the phenomenon that the true match should not only be similar to theStrong similar samples of the probe but also dissimilar to the strong dissimilar samples.
- Image classification using spatial pyramid robust sparse coding47
A new image classification method by spatial pyramid robust sparse coding (SP-RSC), which tries to find the maximum likelihood estimation solution by alternatively optimizing over the codebook and local feature coding parameters, hence is more robust to outliers than traditional sparse coding based methods.
- Intra-View and Inter-View Supervised Correlation Analysis for Multi-View Feature Learning46
This paper proposes a novel multi-view feature learning approach based on intra-view and inter-view supervised correlation analysis (I2SCA), which can explore the useful correlation information of samples within each view and between all views.
- Coupled-View Based Ranking Optimization for Person Re-identification45
A coupled-view based ranking optimization method through cross KNN rank aggregation and graph-based re-ranking to revise the original ranking lists based on the core assumption that the images of the same person should share the similar visual appearance in both global and local views.
- 38
- 37
- Specific Person Retrieval via Incomplete Text Description36
A specific attribute completion is conducted to enrich the original text query and generate a more expressive attribute vector and a pairwise-based metric learning is introduced for completed attribute vectors.
- 36
- Soil type recognition as improved by genetic algorithm-based variable selection using near infrared spectroscopy and partial least squares discriminant analysis35
The results indicate that the GA variable selection technique can significantly improve the accuracy rate of soil type recognition using NIR spectroscopy, suggesting that the proposed methodology is a promising alternative for recognizing soil types using Nir spectra.
- Image classification using Harr-like transformation of local features with coding residuals35
This paper proposes a novel image classification method using Harr-like transformation of local features with additional consideration of coding residuals to reduce the information loss during the local feature encoding process.
- Manifold Projection for Adversarial Defense on Face Recognition34
This paper proposes Adversarial Variational AutoEncoder (A-VAE), a novel framework to tackle off-manifold and on-Manifold adversarial images, and hypothesizes that both off- manifoldand on-manIFold attacks move the image away from the high probability region of image manifold.
- Camera Network Based Person Re-identification by Leveraging Spatial-Temporal Constraint and Multiple Cameras Relations34
Quantitative experiments conducted on TMin and CamNeT datasets have shown that the proposed probabilistic method achieves a better performance to the novel CNPR problem.
- Real-Time Imaging of Intracellular Glutathione Levels Based on a Ratiometric Fluorescent Probe with Extremely Fast Response33
RP-2 was expected to serve as a new fluorescent imaging tool to understand the function of intracellular GSH in the future and had both the largest forward and reverse rate constants thus far.
- 32
- When Pedestrian Detection Meets Nighttime Surveillance: A New Benchmark32
This paper builds a novel pedestrian detection dataset from the nighttime surveillance aspect, and compares the performances of state-of-the-art pedestrian detectors and reveals that the methods cannot solve all the challenging problems of NightSurveillance.
- 31
- Region-Based Interactive Ranking Optimization for Person Re-identification31
This paper proposes a region-based interactive ranking optimization method, to improve the original query result by labeling locally similar and dissimilar image regions in the person re-identification problem, and proves that the ranking optimization algorithm is both an effective and efficient method.
- 30
- 29
- 29
- Multiple Pedestrian Tracking Under Occlusion: A Survey and Outlook28
This survey provides comprehensive review in terms of occlusion scenarios encountered during MPT, and investigates the model robustness of the existing methods in this scenarios, and categorizes existing occlusion handling methods according to the tracking process and detail their pros and cons.
- A Novel Role-Based Movie Scene Segmentation Method28
The main novelty of this approach is that it converts the movie scene segmentation into a movie-script alignment problem and proposes a HMM alignment algorithm to map the script scene structure to the movie content.
- Object categorization in sub-semantic space27
This paper proposes a novel object categorization method by using the sub-semantic space based image representation, which is not only more semantically meaningful than exemplar based representation but also more reliable and resistant to noise than traditional semantic spacebased image representation.
- DNA Sequential Logic Gate Using Two-Ring DNA26
A sequential logic DNA gate was achieved by utilizing the two-ring DNA structure, with the ability to recognize "before" and "after" triggering sequences of DNA signals, to guide future molecular sensors to monitor more complex gene network in biological systems.
- Geometrically Based Linear Iterative Clustering for Quantitative Feature Correspondence26
This paper proposes a novel feature matching algorithm for image collections, capable of providing quantitative depiction to the plausibility of feature matches, and derives the solution of the optimization problem in a simple linear iterative manner, where a k‐means‐type approach is designed to automatically generate consistent feature clusters.
- Face image super-resolution through locality-induced support regression26
A novel face image super-resolution method named Locality-induced Support Regression (LiSR), which defines the support in the LR or HR patch manifold space by incorporating the locality constraint, which can well preserve the manifold structure of the training set.
- 25
- Multi-Level Fusion for Person Re-identification with Incomplete Marks25
This paper studies the problem of Person Re-identification with Incomplete Marks (PRIM), aiming at ranking the persons in the gallery according to both the videos and incomplete marks, and shows that through introducing marks, the retrieval accuracy is significantly improved.
- 25
- 24
- A Survey of Personalized Interior Design24
The main challenge for PID is meeting residents' personalized requirements in terms of both furniture and floor plans, and recent progress is reviewed in five separate but correlated areas, including furniture style analysis, furniture compatibility prediction,floor plan design, floor plan analysis and automatic furniture layout.
- 24
- A Unsupervised Person Re-identification Method Using Model Based Representation and Ranking24
This paper proposes a new unsupervised person re-identification method featured by its competitive accuracy and high efficiency, which is based on model based person image representation and ranking and can achieve over 20% matching rate at Rank 1 on the challenging VIPeR dataset.
- 24
- 23
- 23
- Beyond visual features: A weak semantic image representation using exemplar classifiers for classification23
This paper proposes a simple but effective image representation for image classification, which is denoted as the responses to a set of exemplar image classifiers, which are learned using SVM algorithm to distinguish the image from others in different classes.
- 21
- 20
- 20
- 20
- Correlation Discrepancy Insight Network for Video Re-identification20
A novel correlation discrepancy insight network for video-based person ReID is proposed, which consists of an unsupervised correlation insight model for video purification and a discrepancy description network (DDN) for person representation that outperforms several state-of-the-art methods.
- 19
- Application of A* algorithm in intelligent vehicle path planning19
The results show that the improved A* algorithm has good effect on path planning of intelligent vehicles.
- 17
- 17
- Sparsity-Based Occlusion Handling Method for Person Re-identification17
A sparsity-based patch matching method using a sparse representation model to determine the occlusion state of each image patch, which is further utilized to adjust the weight of patch pairs in the feature matching process.
- SATCount: A scale-aware transformer-based class-agnostic counting framework16
This paper proposes a new scale-aware transformer-based feature fusion module that integrates visual and scale information of exemplar objects and models similarity between samples and queries using cross-attention, and proposes an object counting algorithm based on a feature extraction backbone, a feature fusion module and a density map regression head, called SATCount.
- Improved YOLOv5 infrared tank target detection method under ground background16
A You Only Look Once, Transform Head Squeeze-and-Excitation model is proposed based on the YOLOv5s model, and it is shown that this model can effectively improve the detection performance of infrared tank targets under ground background compared with existing methods.
- 16
- Illumination animating and editing in a single picture using scene structure estimation16
By estimating the scene structure, a novel method to animate the illumination of a single input image is proposed and can effectively process images with complex shadow, multiple lights (including both point and area light sources) and specular spots.
- 15
- Generating video animation from single still image in social media based on intelligent computing15
This paper presents an image animating method for enhancing single still image in social media with virtual realistic and animated motions without prior information that produces visually natural results while guaranteeing motion harmony between active objects and passive objects.
- A novel face super resolution approach for noisy images using contour feature and standard deviation prior15
Different from the existing methods which directly enhance the noisy input image in intensity feature space, the proposed method introduces a contour feature which is robust to noise and which outperforms most classic methods both quantitatively and qualitatively.
- 15
- Image classification using boosted local features with random orientation and location selection15
A novel boosted local features method with random orientation and location selection is proposed to combine the local feature extraction and encoding with classifier training into a unified framework and gradually improve the image classification performance.
- 14
- A Method for Predicting the Remaining Life of Rolling Bearings Based on Multi-Scale Feature Extraction and Attention Mechanism14
A rolling bearing remaining life prediction method based on multi-scale feature extraction and attention mechanism that takes the normalized bearing vibration signal as input and adopts a quadratic function as the RUL prediction label, avoiding identifying the degradation stage start point.
- 14
- 14
- Demo14
DSENSE, a mobile wireless charging and sensing platform using drones to alleviate the pain of energy delivery and data gathering in WRSN is proposed and demonstrated to be effective and efficient.
- 13
- 13
- Person re-identification with multiple similarity probabilities using deep metric learning for efficient smart security applications13
A multiple deep metric learning method empowered by the functionality of person similarity probability measurement powered by the Softmax regression models that can achieve a time complexity as low as O ( n ) , which can be deployed at a large scale on the distributed intelligent surveillance network, with each node maintaining limited computing capabilities.
- 13
- 11
- Foreground Harmonization and Shadow Generation for Composite Image10
An effective model to tackle the problem of foreground lighting-shadow editing of outdoor disharmonious composite images is proposed, using a coarse shadow prediction module (SP) to generate coarse shadows for foreground objects and using the predicted results as prior knowledge to guide the generation of harmony diffusion model.
- Single image de-raining via clique recursive feedback mechanism10
A new rain removal network based on the clique recursive feedback mechanism is proposed, considering the interaction of the feature between different convolution layers, and constructing a residual clique block (RCB) to infer the local information, allowing the network to rectify the model parameters in RCB alternately.
- 10
- Laplacian affine sparse coding with tilt and orientation consistency for image classification10
A Laplacian affine sparse coding algorithm which combines the tilt and orientations of affine local features as well as the dependency among local features is proposed, and tilt and orientation smooth constraints are added into the objective function of sparse coding.
- Biotemplated Diatom Microrobots for Dual-Drug Delivery in Targeted Neuroblastoma Therapy9
This study developed magnetic diatom microrobots using diatom frustules as biotemplates using diatom frustules as biotemplates to demonstrate enhanced drug conjugation and tumor-targeting capabilities in neuroblastoma patients.
- 9
- 9
- 9
- 9
- 9
- 9
- 9
- 8
- CFDiffusion: Controllable Foreground Relighting in Image Compositing via Diffusion Model8
This work designs a harmonization-shadow generator based on a diffusion model to harmonize the foreground and generate shadows concurrently, and proposes a foreground content enhancement module to ensure the complete preservation of foreground content at the insertion location.
- 7
- 7
- Research on lightweight pedestrian detection based on improved YOLOv57
The improved lightweight YOLOv5 algorithm significantly reduces the model size and computational cost while guaranteeing accuracy, which is suitable for deployment in edge devices.
- Combinatorial Library Based on Restriction Enzyme-mediated Modular Assembly7
This approach showed that different DNA fragments without homologous sequences could be randomly assembled to create significant sequence space.
- Vehicle re-identification collaborating visual and temporal-spatial network7
A two-stage strategy is proposed to accomplish vehicle re-identification in realistic surveillance videos using a part-based appearance model fusing multiple visual features to represent the vehicle object and a coarse ranking list is generated by comparing appearance models of the probe and gallery vehicles.
- 7
- 6
- A Hierarchical Deep Video Understanding Method with Shot-Based Instance Search and Large Language Model6
This paper proposes a four-stage method to solve the deep video understanding task, which includes video structuring, shot based instance search, interaction & relation prediction and shot-scene summary & Question Answering (QA) with ChatGPT.
- 6
- Occluded suspect search via channel-guided mechanism6
A channel-guided mechanism (CGM) for occluded suspect search using different body areas have been activated via different channels in convolutional neural networks which can identify suspects using a rule which focuses more on the visible area and focuses less on the occluding area in the specific occlusion pattern.
- Rapid and enzyme-free nucleic acid detection based on exponential hairpin assembly in complex biological fluids6
This method combined the exponential self-assembly of four DNA hairpins and the FRET pair Cy3 and Cy5 to detect nucleic acids at concentrations as low as 10 pM in 15 min in biological fluids.
- Undo the codebook bias by linear transformation for visual applications6
An alternative optimization algorithm is proposed to jointly search for the optimal linear transformation matrixes and the encoding parameters and show the effectiveness of the proposed method on image classification experimental results.
- Personalized Sports Video Customization Using Content and Context Analysis6
A user-participant multiconstraint 0/1 Knapsack model is designed to model the personalized video customization, which can unify both video retrieval and summarization with different fusion parameters.
- 6
- Person-action Instance Search in Story Videos: An Experimental Study5
An indirect evaluation method motivated by the leave-one-out strategy is proposed, which finds an optimal solution surpassing the champion teams in 2020–2021 INS tasks and is validated under various scenarios.
- 5
- 5
- 5
- SIST: Online Scale-Adaptive Object tracking with Stepwise Insight5
This work proposes an online scale-adaptive correlation tracker by collaboratively learning a target-insight model (TIM) for object localization and a scale-insights model (SIM) for progressive refinement, which outperforms several state-of-the-art methods.
- 5
- Person Re-identification Using Data-Driven Metric Adaptation5
This paper proposes a data-driven metric adaptation method that re-exploits the training data with cross-view consistency to adaptively adjust the metric.
- WHU-NERCMS at TRECVID2013:Instance Search Task5
This paper introduces the NERCMS team work at the automatic instance search task of TRECVID 2013, using a stare model to weight query images and several result combination strategies to generate the nal ranking lists.
- Personalized Sports Video Customization for Mobile Devices5
The proposed mobile personalized sports video customization system aims to provide mobile users with interesting video clips according to their personalized preferences and is proved effective by both quantitative and qualitative experiments conducted on five sports matches.
- A Hierarchical Semantics-Matching Approach for Sports Video Annotation5
A hierarchical semantics-matching approach to annotate sports video to link video and text with high-level semantics rather than low-level features and find the optimal video-text alignment based on the integral structure rather than individual conditions.
- 4
- A Virtual Power Plant Load Forecasting Approach Using COM Encoding and BiLSTM-Att-KAN4
A Cyclic Order Mapping (COM) encoding method is introduced, which maps weekly and intraday sequences into continuous ordered variables on the unit circle, thereby effectively preserving load periodic features and demonstrating the effectiveness of the proposed model in enhancing load forecasting performance for VPP applications.
- 4
- 4
- 4
- 4
- 4
- 4
- WHU-NERCMS at TRECVID2016: Instance Search Task.4
This report introduces the work at the instance search task of TRECVID 2016, which asks systems to retrieve specific persons in specific locations and exploits e-ective face recognition method, which detect faces by a scale-adaptive deconvolutional regression network and then recognize detection results by a deep embedding network.
- Beyond visual word ambiguity: Weighted local feature encoding with governing region4
A novel image representation method is proposed by going one step beyond visual word ambiguity and consider the governing regions of visual words and a weighted feature sign search algorithm is proposed to solve the problem.
- Beyond bag of words4
A novel image representation method in the sub-semantic space based representation of images that is not only more semantically meaningful but also more reliable and resistant to noise.
- 3
- XLNet-CRF: Efficient Named Entity Recognition for Cyber Threat Intelligence with Permutation Language Modeling3
XLNet-CRF is proposed, a hybrid framework that combines permutation-based language modeling with structured prediction using Conditional Random Fields (CRF) to enhance Named Entity Recognition (NER) in cybersecurity contexts and significantly surpassing strong BERT-based baselines in both accuracy and robustness.
- 3
- 3
- An improved YOLOv5 algorithm for obscured target recognition3
SENet mechanism in YOLOv5 can make the network better to learn the locations in the images that need attention and show that the improved Yolov5 algorithm has better results for target detection.
- 3
- 3
- Tool life prediction based on multi-source feature PSO-SVR neural network3
A tool life prediction method based on multi-source feature PSO-SVR neural network is proposed, monitoring and collecting the current and vibration signals of the tool during the machining process of CNC machine tools to predict the remaining useful life(rul) of the tools.
- 3
- A Multi-scale Triplet Deep Convolutional Neural Network for Person Re-identification3
Experimental results on two public person re-identification datasets demonstrate that the proposed MVT-CNN method outperforms most of the existing feature learning algorithms by 8%–10% at rank@1 in term of the cumulative matching curve (CMC) criterion.
- Identity Recognition Algorithm Using Improved Gabor Feature Selection of Gait Energy Image3
The kernel Fisher analysis combined with kernel matrix is proposed to select dominant features of gait energy image and the nearest neighbor classifier based on whitened cosine distance is used to discriminate different gait patterns.
- P-S Instance Retrieval via Early Elimination and Late Expansion3
An early elimination and late expansion method is proposed to improve the accuracy of P-S Instance Retrieval, and experimental results on the large-scale TRECVID INS dataset demonstrate the effectiveness of the proposed method.
- Person Re-Identification via Multiple Coarse-to-Fine Deep Metrics3
This paper proposes a coarse-to-fine deep metric learning method equipped with multiple different Stacked Auto-Encoder (SAE) networks and classification networks that simulate the information processing of the brain’s visual system to recognize the character of objects.
- Person Re-identification via Attribute Confidence and Saliency3
A novel person re-identification method that epitomizes the diversities of different attributes as two folds: the attribute confidence which denotes the descriptive power, and the attribute saliency which expresses the discriminative power is proposed.
- Undoing the codebook bias by linear transformation with sparsity and F-norm constraints for image classification3
The sparsity constraints and the F-norm of the transformation matrix are used in this paper to propose an alternative optimization algorithm to jointly search for the optimal linear transformation matrixes and the encoding parameters.
- Person Re-identification by Smooth Metric Learning3
This paper proposes to replace hinge loss function with the logistic loss function, which transforms LMNN to a smooth unconstrained convex optimization problem easily solved with gradient descent algorithm and designs a stochastic sampling scheme to accelerate the optimization process of the above problem with randomly selected training samples.
- 3
- 3
- 3
- 2
- 2
- 2
- Quantum Interference-Inspired Who-What-Where Composite-Semantics Instance Search for Story Videos2
A Quantum Interference Partial Decomposition (QIPD) method is proposed to model the diverse influences of semantic overlap from 2W to 3W INS, which incorporates two core modules, i.e., semantic interference and temporal interference.
- 2
- 2
- 2
- 2
- 2
- WHU-NERCMS at TRECVID2015: Instance Search Task.2
The main improvement in the strategies is Query Expansion With Adjacen-t Shots, which aims to retrieve more precisely in the adjacent shots of initial top-k results.
- 2
- WHU-NERCMS at TRECVID2014: Instance Search Task.2
The work at the automatic instance search task of TRECVID 2014 is divided into two parts: first part is object retrieval based on BOW, and secondly, several optimization strategies are adopted to improve the initial results.
- Person re‐identification based on contextual characteristic2
An efficient contextual characteristic is proposed for person re-identification that contains similarities of both k-nearest and ḱ-farthest neighbours between the probe and the gallery, and combines it with Mahalanobis distance for ranking every gallery image more accurately.
- 1
- 1
- IDRetracor: Towards Visual Forensics against Malicious Face Swapping1
This work proposes IDRetracor, a model designed to recover arbitrary original target identities from fake faces generated by various face swapping techniques, and introduces Mapping-Aware Convolutions (MACs), which consist of multiple dynamically combined kernels guided by the mapping resolver to adaptively handle diverse face swapping patterns.
- InstructSee: Instruction-Aware and Feedback-Driven Multimodal Retrieval with Dynamic Query Generation1
A novel cross-modal representation learning framework that incorporates an instruction-aware dynamic query generation mechanism, augmented by the semantic reasoning capabilities of large language models (LLMs) to effectively infer and adapt to implicit retrieval intent is proposed.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Multi-Pedestrian Detection from Effective Proposal in Crowd Scene1
A novel MP detection method is presented that introduces bottom-up region proposals (which are called effective proposals) into the MP detection process, from which, high detection rate as well as low computation complexity can be simultaneously achieved.
- 1
- Saliency-Based Deformable Model for Pedestrian Detection1
In SBDM method, it is presented that, besides the local features, the saliency in the image provides important constraints that are not yet well utilized, and a probabilistic framework is proposed to model the relationship between Saliency detection and the feature Deformable Model via a Bayesian rule to detect pedestrians in the still image.
- Histograms of Salience for Pedestrian Detection1
A salience representation for pedestrian detection, Histograms-Of-Salience (HOS) is proposed and investigated, extracted saliency map learned from data by using Histogram Based Contrast, and aggregate salient value and oriented gradients to form local HOS.
- Fine-Grained Image Classification Using Color Exemplar Classifiers1
This paper proposes a novel fine-grained image classification method by using color exemplar classifiers, which first decompose it into multiple color channels to take advantage of the color information.
- Morphological variations and discriminant analysis of different geographical populations of Scapharca broughtonii Schrenck1
The multivariate morphometrics based on 9 morphological characters were used to investigate the morphological variations among the four different geographical populations of S.broughtonii and can be used in discrimination of different geographical population and sustainable utilization and protection of the breed.
- Personalized Sports Video Customization Based on Multi-modal Analysis for Mobile Devices1
A mobile personalized sports video customization system for mobile users by using a novel approach based on the B/S architecture, which adopts basketball and football games as its initial sports genres.
- 1
- 1
- 1
- 1
- DUIE: Depth-Aware Underwater Image Enhancement via Spatial Prior Modeling–
A Depth-Aware Underwater Image Enhancement (DUIE) framework that uses relative depth as a spatial prior for continuous enhancement control and combines a physics-inspired depth cue, interpretable statistical candidates, and learned local refinement without explicitly inverting a complete underwater image formation model is proposed.
- –
- StoryVideoQA: Scaling Deep Video Understanding with a Large-Scale, Multi-Genre and Auto-Generated Dataset–
PlotTree is proposed, a novel video understanding agent, re-organizing long-range video content into a hierarchical plot structure, enabling efficient storyline reasoning on StoryVideoQA.
- –
- –
- A Robust 3D CNN with Pyramidal Attention for Spatiotemporal Gait Recognition–
A novel three-dimensional convolutional architecture named Robust Spatiotemporal 3D Convolutional Neural Network (RST3D), which jointly captures spatial and temporal correlations throughout gait sequences and achieves superior performance under challenging conditions such as view variation, clothing changes, and occlusion.
- –
- –
- –
- –
- –
- Bias-Rectified Multi-way Learning with Data Augmentation for Implicit Discourse Relation Recognition–
- –
- –
- –
- –
- –
- –
- Confidence-Aware Active Feedback for Interactive Instance Search–
A confidence-aware active feedback method that is specifically designed for online RF in interactive INS tasks, Inspired by the explicit difficulty modeling scheme in self-paced learning, CAAF utilizes a pairwise manifold ranking loss to evaluate the ranking confidence of each unlabeled sample.
- –
- Moving object recognition with optical flow-based fusion of thermal and visible videos–
This paper presents a two-stage approach to moving object recognition by jointly utilizing the thermal and visible videos and designs a deep network to recognize the categories of the object proposals generated in the first stage and thus obtain the recognition results.
- –
- –
- –
- Transferring fashion to surveillance with weak labels–
This paper addresses the problem of automatic clothing parsing in surveillance images using the information from user-generated tags, such as “jeans” and “T-shirt" using a weakly supervised transfer learning method.
- Video-Based Person Re-Identification With Unregulated Sequences–
An adaptive unregulated sub sequence detection and refinement method is proposed to purify original video sequence and obtain a more effective and discriminative feature representation for video-based person re-id.
- –
- –
- A New Non-local Mean Method for the Image Denoising of Coal Dust–
A new non-local mean image denoising algorithm is designed by using the Gaussian kernel function of weighted of cosine coefficient, and applied to the weighting coefficient calculation and the experimental results show that the Denoising performance of the algorithm is better than the traditional algorithm and can better preserve the details of coal dust image.
- –
- –
- –
- –
- –
- Behavior analysis of video object in complicated background–
The theory and method on the behavior analysis of video object in reality scenes put forward by this project have broad application prospect and important practical significance in the security, terrorism, military and many other fields.
- Real-Life Voice Activity Detection Based on Audio-Visual Alignment–
This work proposes a novel method to detect voice activity by audio-visual alignment, which uses Needleman-Wunsch algorithm to align two different sequences.
- –
- –
- A New Method for Objective Communication Speech Quality Evaluation–
A new speech feature extraction method is proposed on the base of MFCC using Gammatone filter bank instead of triangle filter bank, and RASTA filter to further process the results of the previous step; the experimental results show that the proposed algorithm is effectively in speech evaluation.
- –
- –
- AN OPEN STRUCTURE OF TESTING PROGRAM DEVELOPMENT ENVIRONMENT BASED ON REUSE IN THE LARGE–
An open architecture of a data oriented incremental testing software development environment (TESSDE) is constructed based on the technology of software reuse, which can lead to an open system architecture and improve the incrementality of the resulted software.
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.