Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks230 from public data
- Deep Image Matting513
A novel deep learning based algorithm that can tackle image matting problems when an image has similar foreground and background colors or complicated textures and evaluation results demonstrate the superiority of this algorithm over previous methods.
- Deep Interactive Object Selection494
This paper presents a novel deep-learning-based algorithm which has much better understanding of objectness and can reduce user interactions to just a few clicks and is superior to all existing interactive object selection approaches.
- 299
- 217
- 173
- 172
- Streamlined Dense Video Captioning171
A novel dense video captioning framework is proposed, which models temporal dependency across events in a video explicitly and leverages visual and linguistic context from prior events for coherent storytelling.
- Deep GrabCut for Object Selection141
A novel segmentation approach that uses a rectangle as a soft constraint by transforming it into an Euclidean distance map that gets accurate segmentation results given sloppy rectangles while being general for both interactive segmentation and instance segmentation.
- 128
- 127
- 120
- 114
- 109
- Scene graph captioner: Image captioning based on structural visual representation97
This work proposes a novel framework to embed a scene graph into the structural representation, which captures the semantic concepts and the graph topology and develops the scene-graph-driven method to generate the attention graph.
- 95
- 94
- 64
- Single/multi-view human action recognition via regularized multi-task learning57
The pyramid partwise bag of words (PPBoW) representation is proposed which implicitly encodes both local visual characteristics and human body structure and can achieve competing performance against the state-of-the-art methods for human action recognition on KTH and MV-TJU.
- Shape and View Independent Reflectance Map from Multiple Views57
The View Independent Reflectance Map (VIRM), which is a representation of the joint effect of the diffuse+specular Bidirectional Reflectance Distribution Function (BRDF) and the environment illumination, is used to model the reflectance of an object.
- 54
- 44
- 44
- Learning by Planning: Language-Guided Global Image Editing42
A novel operation planning algorithm to generate possible editing sequences from the target image as pseudo ground truth is proposed and compared on the newly collected MA5k-Req dataset and GIER dataset to show the advantages of these methods.
- Asymptotical stabilisability of networked control systems: time-delay switched system approach40
The authors focus on investigating the asymptotical stabilisability of networked control systems (NCSs) with communication constraints, and find that the behaviour of NCSs can be studied by considering time-delay-switched systems without augmenting the system dimension.
- A Benchmark and Baseline for Language-Driven Image Editing39
This work presents a new language-driven image editing dataset that supports both local and global editing with editing operation and mask annotations and proposes a baseline method that fully utilizes the annotation to solve this problem.
- 38
- 38
- 38
- Hierarchical & multimodal video captioning: Discovering and transferring multimodal knowledge for vision to language36
A Hierarchical & Multimodal Video Caption (HMVC) model is proposed to jointly learn the dynamics within both visual and textual modalities for video caption task, which infers an arbitrary length sentence according to the input video with arbitrary number of frames.
- 35
- End-To-End Time-Lapse Video Synthesis From a Single Outdoor Image35
An end-to-end solution to synthesize a time-lapse video from a single outdoor image using deep neural networks and a multi-frame joint conditional generation framework to effectively learn the correlation between the illumination change of an outdoor scene and the time of the day.
- Co-training and visualizing sentiment evolvement for tweet events33
A "river" graph is proposed to visualize the intensity and evolvement of sentiment on an event, which demonstrates the intensity by both color gradient and opinion labels, and the ups and downs of confronting opinions by the river flow.
- Discrete symbiotic organism search with excellence coefficients and self-escape for traveling salesman problem30
ECSDSOS shows competitive optimization capacities for computing better solutions for TSP than basic DSOS and some other heuristic algorithms.
- Recognition of In-air Handwritten Chinese Character Based on Leap Motion Controller30
This paper proposes a novel 3D interaction method by recognizing Chinese character written in the air by combining the directional feature and direction-change feature using the Leap Motion Controller.
- Multi-Level Policy and Reward Reinforcement Learning for Image Captioning29
This work proposes a novel multi-level policy and reward RL framework for image captioning that contains a Multi-Level Policy Network that can adaptively fuse the word- level policy and the sentence-levelpolicy for the word generation and a guidance term to bridge thepolicy and the reward for RL optimization.
- 28
- Deep neural network based i-vector mapping for speaker verification using short utterances27
This paper compares two state-of-the-art universal background model (UBM) training methods for i-vector modeling using full-length and short utterance evaluation tasks and proposes two novel nonlinear mapping methods which train DNN models to map the i-vectors extracted from short utterances to their corresponding long-utterance i-VEctors.
- Multi-modal & Multi-view & Interactive Benchmark Dataset for Human Action Recognition27
This paper introduces a Multi-modal & Multi-view & Interactive (M2I) dataset, which is designed for the evaluation of the performances of human action recognition under multi-view scenario, and evaluates three representative methods for the single-view, cross view, and cross domain humanaction recognition on this dataset.
- 25
- Distributed Incentives and Digital Twin for Resource Allocation in air-assisted Internet of Vehicles25
- 24
- Hierarchical Deep Neural Network for Image Captioning24
A hierarchical deep neural network is proposed, which consists of the bottom layer and the top layer, which extracts the visual and high-level semantic information from image and detected regions, respectively, while the latter integrates both of them with adaptive attention mechanism for the caption generation.
- Multi-Modal Validation and Domain Interaction Learning for Knowledge-Based Visual Question Answering21
- 20
- 16
- 16
- Part-Aware Interactive Learning for Scene Graph Generation16
A part-aware interactive learning method that can highlight the subject-related object parts and the object-related subject parts during inter-object interactive learning and demonstrate the superiority of the method against the state of the arts on Visual Genome is proposed.
- 16
- 3D Model classification based on few-shot learning16
A dual-meta-learner model based on LSTM is proposed to learn the exact optimization algorithm used to train another two learner neural network classifier in the few-shot regime to achieve state-of-the-art performance.
- 16
- A Hybrid Chaos Optimization Method and Its Application16
Using the properties of stochastic, ergodicity and regularity of chaos, and the intelligence of tabu search, a hybrid optimization algorithm is proposed that shows the efficiency of CSTA is much higher than some other algorithms.
- Mask and Predict15
A novel multi-step reasoning manner for SGG is proposed, which breaks SGG into two explicit learning stages, including intuitive training stage (ITS) and rationalTraining stage (RTS), yielding an intuitive scene graph.
- Triangle-Reward Reinforcement Learning15
This work argues that the XE objective is not sensitive to visual-linguistic alignment, which cannot discriminately penalize the semantic inconsistency and shrink the context gap, and proposes the Triangle-Reward Reinforcement Learning (TRRL) method, which demonstrates the superiority of TRRL on the MSCOCO dataset.
- 15
- 14
- 13
- FanoutNet: A Neuralized PCB Fanout Automation Method Using Deep Reinforcement Learning13
This work is the first in the literature to propose a neuralized PCB fanout method by deep reinforcement learning, and achieves 100% routability in all industrial cases and improves wire length by an average of 6.8%, which makes a significant improvement compared with the state-of-the-art methods.
- 13
- A Self-Organized Reciprocal Decision Approach for Sensing Coverage with Multi-UAV Swarms13
This paper tackles the problem of sensing coverage for multiple Unmanned Aerial Vehicles with an approach that takes into account the reciprocal between neighboring UAVs to reduce the oscillation of their trajectories.
- 12
- 12
- Regularity-constrained floorplanning for multi-core processors12
This work proposes a simple and effective technique for encoding the regularity constraint in sequence-pairs in the context of multi-core processor designs and investigates how to enforce regularity constraints in a simulated-annealing based floorplanner.
- 12
- 12
- 11
- 11
- 11
- On-line Sample Generation for In-air Written Chinese Character Recognition Based on Leap Motion Controller11
An on-line sample generation method to enlarge the number of training instances in an automatic synthesis way by distorting the inner structure of character patterns and demonstrating higher recognition accuracies and lower computational cost.
- 11
- A discriminative classifier for in-air handwritten Chinese characters recognition11
This paper tackles the problem caused by similar characters using two discriminative analysis techniques, exploiting the learning vector quantization (LVQ) and the adaptive discrim inative locality alignment (ADLA) to distinguish candidate character class among similar character classes.
- SMTPD: A New Benchmark for Temporal Prediction of Social Media Popularity10
It is verified that temporal alignment and early popularity play crucial roles in social media popularity prediction for not only deepening the understanding of temporal dynamics of popularity in social media but also offering a suggestion about developing more effective prediction models in this field.
- 10
- 10
- 9
- 9
- 9
- Multi-Hop Quantum Communication Based on Greenberger-Horne-Zeilinger States and Teleportation9
This paper proposes a faithful multi-hop quantum communication scheme that shows advantages in quantum communication delay and transfers the quantum state of a physical system to the receiver independently.
- Multi-guiding long short-term memory for video captioning9
A novel cell-wise ensemble, where the weight matrix of each cell of LSTM is extended to be a set of attribute-dependent and attention-dependent weight matrices, by which the guiders induce each cell optimization over time.
- Construction of a knee osteoarthritis diagnostic system based on X-ray image processing9
The experimental results show that this detection technique can effectively detect and describe the features of knee osteoarthritis, which can be used as a tool for clinical diagnosis.
- 8
- Multimodal information fusion based on LSTM for 3D model retrieval8
A multimodal feature fusion method based on the LSTM network to solve the problem of 3D model retrieval skill shortage and demonstrates the superiority of this approach.
- 8
- 8
- 8
- 7
- 7
- Cross-Modal Coherence-Enhanced Feedback Prompting for News Captioning7
This work proposes the novel cross-modal coherence-enhanced feedback prompting method to clarify the crucial elements that align closely with the visual content for news captioning.
- A comprehensive survey on deep-learning-based visual captioning7
This paper presents a comprehensive survey of existing deep learning-based visual captioning methods, relying on the adopted mechanism and technique to narrow the semantic gap, and divides visual captioned methods into various groups.
- Multi-type decision fusion network for visual Q&A7
The novel multi-type decision fusion model, which can adaptively leverage object- and relation-level decision networks to infer the answer, is introduced.
- Filter Sampling and Combination CNN (FSC-CNN): A Compact CNN Model for Small-footprint ASR Acoustic Modeling Using Raw Waveforms7
A novel method to generate CNN parameters by first sampling from a low-dimensional parameter space and then using a trainable scalar vector to perform a linear combination, which is feasible for small-footprint ASR and outperforms a CNN model with a similar number of parameters.
- Graph cut based clustering for cognitive radio ad hoc networks without common control channels7
A novel distributed clustering algorithm for CRAHNs, where neither a dedicated common control channel (CCC) nor prior topology information is required, and a proactive cluster maintenance mechanism is proposed to reduce the interferences caused by PU activities.
- A kind of PAPR reduction method based on pruning WPM and PTS technology7
Simulation results show that when Complementary Cumulative Distribution Function (CCDF) is 10−3, PTS based on pruning WPM compared with PTS technique and pruning technique has improved about 1 dB and 1.5 dB, which will not affect the system’s Bit Error Rate (BER) performance in the wavelet packet multicarrier system.
- 6
- 6
- A Hybrid Particle Swarm Optimization Method for Traveling Salesman Problem6
The results show that the much better approximations are searched with the hybrid PSO, and the convergence rate is also faster than the traditional PSO under the same conditions.
- An LTL-Based Motion and Action Dynamic Planning Method for Autonomous Robot**This work was supported by National Natural Science Foundation of China (61403410).6
A method to dynamically generate motion and action sequences for a robot based on LTL task specification and a weighted finite transition system is constructed to capture the robot’s full functionality and the knowledge of environment.
- 5
- 5
- 5
- Rule-Driven News Captioning5
This paper proposes the rule-driven news captioning method, which can generate image descriptions following designated rule signal and injects this semantic rule into the large-scale pre-trained model, BART, with the prefix-tuning strategy, where multiple encoder layers are embedded with news-aware semantic rule.
- SMPC: boosting social media popularity prediction with caption5
This paper proposes social media popularity prediction with caption (SMPC), a novel architecture that integrates the caption as the global representation into the existing multi-model-feature-based popularity prediction method.
- Multi-stage reasoning on introspecting and revising bias for visual question answering5
A bias discrimination module is proposed to distinguish good and bad bias indiscriminately in existing debiasing methods and helps with feature updating for both image regions and question words, thereby facilitating more accurate and comprehensive reasoning about both the image content and question.
- 5
- Cross-Media Body-Part Attention Network for Image-to-Video Person Re-Identification5
A Cross-media Body-part Attention Network (CBAN) for image-to-video person re-id, which can extract the cross-media body part attention features from images/videos, and simultaneously ignore the useless information in the background by using a part attention mechanism.
- Constitutive Relation Research of Q235 Steel Based on Support Vector Machine5
The research shows that constitutive relation equations based on experimental data and forecast data could show a complex nonlinear relationship between mechanical properties and temperature, and comes the conclusion that prediction model via SVM have the advantage of less relative error, fewer calculation, high goodness of fit.
- Ontological Random Forests for Image Classification5
The authors propose an ontological random forest algorithm where the of decision trees are determined by semantic relations among categories and hierarchical features are automatically learned by multiple-instance learning to capture visual dissimilarities at different concept levels.
- An encryption scheme based on phase-shifting digital holography and amplitude-phase disturbance5
A mathematical model is developed and a discrete formula is given based on the scheme, which makes it easy to implement the scheme in computer programming and demonstrates a good capability of anti-noise and anti-shear robustness.
- Model Can Be Subtle: Two Important Mechanisms for Social Media Popularity Prediction4
This article divides social network users into two categories and finds a dilemma in existing models and proposes the knowledge embedding mechanism to revise the popularity scores in existing models by fusing the statistical frequency over multi-type data.
- 4
- 4
- Robust water meter reading recognition method for complex scenes4
In this method, the traditional water meter recognition method has been improved so that it can handle more variable real-world scenarios and a data augmentation method is proposed for half-word character recognition that is difficult to handle in the past, which greatly improve network performance.
- Segments-Based 3D ConvNet for Action Recognition4
This paper proposes Segments-based 3D ConvNet (S3D) to integrate both long-term and short-term temporal dynamics, and utilizes 3D ResNet without temporal downsampling to capture short-range video contents.
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 3
- 3
- 3
- 3
- Closed-loop reasoning with graph-aware dense interaction for visual dialog3
This work proposes closed-loop reasoning with graph-aware dense interaction, aiming to discover cues through the dynamic structure of graph and leverage it to benefit dialog and image features.
- Deep triplet-group network by exploiting symmetric and asymmetric information for person reidentification3
To utilize discriminative information in triplets and make impostor and its congeners more clustering, oversymmetric and overasymmetric relationships are designed and applied to triplet and impostors’ congeners to train the deep triplet-group network with original individual images rather than handcrafted features.
- 3
- 3
- Thermal aware floorplanning using Gauss-Seidel method3
A class of sparse linear systems in iterative algorithm is defined and the Gauss-Seidel method is applied to the thermal model of floorplan of VLSI physical design to show theory analysis and extendibility.
- 3
- 2
- 2
- 2
- MCE-HGCN: Heterogeneous Graph Convolution Network for Analog IC Matching Constraints Extraction2
The heterogeneous matching constraint extraction graph neural network (MCE-HGCN) excels at extracting matching constraints from various analog circuits and processes, offering valuable insights for placement guidance and enhancing the efficiency of analog IC layout design.
- 2
- 2
- 2
- 2
- 2
- Temporal Logic Control Synthesis for Distributed Multi-Agent Cooperative Tasking2
The proposed method first synthesizes satisfiable global controllers for uncontrolled system, such that the global specification is met; and searches for a decomposable global controller among them over a maximum synchronization scheme; finally, further refining the synchronization scheme to obtain decomposed distributed controllers.
- 2
- 2
- 2
- New Progress in Mineral Processing Automation2
The progress of on-line analyzing and measuring technology, the process optimizing controlling technology, and the plant modeling simulating technology in the recent 5 years is summarized.
- 2
- 2
- The Geometry Optimization of Argon Atom Clusters Using Differential Evolution Algorithm2
Experimental results show that the exact global optimal configuration of argon clusters with atom number N≤ 16 can be found in a reasonable computing time, and approximate optimization can be obtained for clusters with N=30.
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Synthesis of Asynchronous Control and Synchronization for Multi-Robot System under Finite Horizon Temporal Task1
An automated control synthesis method is proposed, which generates provably correct distributed controllers by synthesizing satisfiable global controllers for the parallel composition of the uncontrolled system, such that the global specification is met.
- Degradation analysis and optimization of temperature effect on MEMRISTOR-based Neural Network Accelerators by electro-thermal simulation1
This work establishes an electro-thermal simulation platform to evaluate the temperature impact on memristor -based NN accelerators and applies a temperature-aware NN weight mapping scheme to the most temperature-sensitive layer and achieves 28.89% improvement in computation accuracy.
- Algorithm Modeling of Smile Intensity Estimation Using a Spatial Attention Convolutional Neural Network1
A self-weighted deep convolutional neural network architecture for smiles intensity estimation using graphics processing unit achieves better accuracy by just using CK+ smile images than state-of-the-art techniques.
- 1
- 1
- Urgent image-to-video person reidentification by cross-media transfer cycle generative adversarial networks1
A cross-media transfer cycle generative adversarial networks (CTC-GAN) network is proposed to solve the urgent image-to-video person reidentification (UIVPR) problem, and experimental results demonstrate that the method can solve UIVPR effectively.
- 1
- 1
- A Novel Colored Time Petri Net Model for Policy Management and Context Awareness Procedure in MANET1
A novel Colored Time Petri Net (CTPN) Model is proposed to describe the procedure of policy management and context awareness and a simplified support graph of context awareness procedure is proposed in order to illustrate the effectiveness of the method.
- Thermal Floorplan Base on Conjugate Gradient Solver in HotSpot1
A class of dummy sparse linear systems is defined in iterative thermal floorplan algorithm, the iterative methods for linear system can be extended to apply to other iterative framework algorithm.
- 1
- 1
- 1
- Design and Realization of a Digital Filter in a Real-Time Audio Signal Acquisition System1
The real-time signal acquisition and sending and displaying system based on TMS320VC5416 DSP and TLV320AIC10 A/D,D/A chip has been successfully applied to the detection of underground pipeline leakage, and has achieved good results.
- 1
- 1
- Design of Streaming Media Player of IPTV STB1
This paper introduced the design of streaming media player of IPTV STB, which was connected with middleware through message queuing and at the same time connected with player module through the entrance of the player.
- A Tentative Approach to Minimal Reducts by Combining Several Algorithms1
By binary search in the area, the minimal reducts can be gotten quickly and doubtlessly and the least condition attributes to describe the system’s classified characteristics can be known.
- A Simple Reduction Analysis and Algorithm Using Rough Sets1
It can be seen that an information system has the similar characteristics relative to a relation database table and can be analyzed using its data table structure and a simple reduction analysis can be completed and get useful reduction information.
- 1
- 1
- –
- –
- –
- –
- HybridNet: a deep hybrid CNN-Transformer network for accurate VLSI congestion prediction–
Experimental results validate that the proposed HybridNet approach consistently outperforms state-of-the-art methods, proving its effectiveness and scalability for early routing congestion prediction.
- –
- How to Understand Named Entities: Using Commonsense for News Captioning–
Commonsense knowledge is exploited to understand named entities for news captioning by correlating the news content with commonsense in the wild, which helps an agent to distinguish semantically similar named entities and describe named entities using words outside of training corpora.
- PCB routing on unstructured meshes with conflict-based search–
This research extends robot path planning to the EDA domain, providing new perspectives and solutions for PCB routing by using an enhanced conflict-based search algorithm applied to Delaunay grids, termed UGPCB-CBS.
- –
- –
- –
- –
- Design of troubleshooting tool for IOT terminal access problems–
A troubleshooting tool based on the structure of the IOT management platform that can simulate the actual situation, to help users locate the fault location and find solutions efficiently and accurately and can shorten the time for troubleshooting and significantly improve the work efficiency.
- Research on terminal security access of the power Internet of Things–
A secure access scheme for electric Internet of Things terminals, which includes three modules: secure SDK, terminal encryption management module and agent component, which is used to realize terminal security encryption and decryption.
- A DTI registration algorithm based on spectral representation–
An explicit representation of the solution for a DTI registration model is presented and a new DTI registration algorithm is induced to show the efficiency of the proposed algorithm in DTI registration.
- –
- A Legalization Algorithm for Heterogeneous FPGAs–
The algorithm in this paper is improved on the abacus algorithm and uses subspaces to evaluate the degree of congestion of cells and sort them, and use optional columns to filter to achieve a reasonable allocation of resources.
- –
- Lookup Table Merging for FPGA Technology Mapping–
In this paper, the dual-output LUTs are merged in the technology mapping stage, and a new method of merging single-outputLUTs into dual- outputs is proposed, and the experimental results show that the algorithm used in this paper has an average reduction of 10.21 % in the number of Luts without worsening delay.
- –
- Synthesis method of strategy for multi-robot systems from local automatons under probabilistic model checking–
The results show that the method can synthesize correct strategies online for multi-robot systems which has no restriction on the LTL operators and reduce the computational burden brought by the automaton-based approach.
- –
- –
- –
- –
- –
- –
- A perception based bit allocation for 3DTV broadcasting–
A perceptually improved bit allocation scheme is proposed by employing a more accurate distortion model inspired by a state-of-the-art perceptual video quality metric that achieves better perceptual rate-distortion performance and guarantees the accuracy of rate control well.
- A Novel Social Context-Aware Data Forwarding Scheme in Mobile Social Networks–
A new metric for data forwarding based on social context information is introduced, in which node’s social context Information is used to calculate the encounter utility between a node and destination, and the social relationship of network nodes is usedto calculate the betweenness centrality utility of a node.
- An adaptive correction algorithm of non-linear distorted document–
Experimental results show that the method proposed is effective and robust in correcting the curl of text lines in various degrees and directions.
- –
- Text Document Classification Using Support Vector Machine with Feature Selection Using Singular Value Decomposition–
The study shows that Support Vector Machine method with choosing features by singular value decomposition (SVD) method is better than other methods and decision tree.
- –
- –
- –
- Design on Simulation of OADM Based on AWG–
A systematic simulation of the add-drop multiplexer(OADM) based on 4x4 AWG was made by using optical simulation soft Optisystem, and the upload and download of optical signals in transmission channel were achieved.
- –
- –
- –
- Data Organization of the Tianjin City Components Status Investigation System–
The purpose, process, application value of the Tianjin city components status investigation system is discussed in order to promote economy in the application and development in the GIS industry.
- –
- –
- –
- –
- Application of OpenGL in Automatization Frame–
The meshy frame was rotatable,translatable,amplificatory, and contractible, and adopt optimized OpenGL pick up elements, that is convenient for the user to observe the press endured by the sensors.
- Performance and Low Power Driven Floorplanning–
The fuzzy technique and the simulated annealing algorithm are employed aiming at low power dissipation and high performance and are evaluated using selected MCNC benchmark circuits.
- –
- –
- –
- –
- A Game-Theoretic Account of Discourse Anaphora–
Kibble (2003c)’s game-theoretic account of discourse anaphora is introduced, its merits and defects are pointed out, and a tentative solution to the overly idealized standards within its framework is proposed.
- Palmprint Recognition Based on Template Learning–
The experiment results suggest that the efficient template learning algorithm used for feature extraction of palmprint solves the drawback of preprocessing in a large extent and achieves high recognition rate.
- –
- A Special Pilpeline Architecture for Morphologic Image Analysis and Its Application in Fast Thining of Character.–
A new pipeline architecture for morphological image analysis is proposed which is high in speed, efficient in computation and flexible in doing different kinds of morphological transformations.
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.