Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks130 from public data
- 1,146
- 522
- Segment Anything Is Not Always Perfect: An Investigation of SAM on Different Real-world Applications289
This work analyzes and discusses the benefits and limitations of SAM, while also presenting an outlook on its future development in segmentation tasks, to give a comprehensive understanding of SAM’s practical applications.
- 278
- Video-of-Thought: Step-by-Step Video Reasoning from Perception to Cognition222
This is the first attempt at successfully implementing the Chain-of-Thought technique for achieving human-level video reasoning, where it shows great potential in extending it to a wider range of video understanding scenarios.
- 203
- Accurate RGB-D Salient Object Detection via Collaborative Learning190
A novel collaborative learning framework where edge, depth and saliency are leveraged in a more efficient way, which solves those problems tactfully and makes the model more lightweight, faster and more versatile.
- Video as Conditional Graph Hierarchy for Multi-Granular Question Answering154
This work proposes to model video as a conditional graph hierarchy which weaves together visual facts of different granularity in a level-wise manner, with the guidance of corresponding textual cues to align with the multi-granular essence of linguistic concepts in language queries.
- 153
- 132
- Fine-Grained Scene Graph Generation with Data Transfer127
This work proposes a novel Internal and External Data Transfer (IETrans) method, which can be applied in a plug-and-play fashion and expanded to large SGG with 1,807 predicate classes, which tries to relieve the data distribution problem by automatically creating an enhanced dataset that provides more sufficient and coherent annotations for all predicates.
- 120
- VPGTrans: Transfer Visual Prompt Generator across LLMs118
This work investigates the VPG transferability across LLMs, and develops a two-stage transfer framework named VPGTrans, which is simple yet highly effective and demonstrated to significantly speed up the transfer learning process without compromising performance.
- 118
- Memory-oriented Decoder for Light Field Salient Object Detection108
A deep-learning-based method where a novel memory-oriented decoder is tailored for light field saliency detection and deeply explore and comprehensively exploit internal correlation of focal slices for accurate prediction by designing feature fusion and integration mechanisms.
- Composed Image Retrieval with Text Feedback via Multi-grained Uncertainty Regularization94
A unified learning approach to simultaneously modeling the coarse- and fine-grained retrieval by considering the multi-grained uncertainty is introduced, which prevents the model from pushing away potential candidates in the early stage, and thus improves the recall rate.
- 90
- 82
- 79
- Learning Frequency-Adapted Vision Foundation Model for Domain Generalized Semantic Segmentation66
A novel Frequency-Adapted (FADA) learning scheme to separately tackle the content and style information by frequency tokens throughout the learning process, which is able to stabilize the scene content and eliminate its impact to DGSS.
- 66
- Visually-Prompted Language Model for Fine-Grained Scene Graph Generation in an Open World63
A Cross-modal prediCate boosting (CaCao) framework, where a visually-prompted language model is learned to generate diverse fine-grained predicates in a low-resource way for open-world predicate scene graph generation (Epic).
- 57
- Learning Spectral-Decomposited Tokens for Domain Generalized Semantic Segmentation56
A novel Spectral-dEcomposed Token (SET) learning framework to advance the frontier and develops an attention optimization method to bridge the gap between style-affected representation and static tokens during inference.
- Promoting Saliency From Depth: Deep Unsupervised RGB-D Saliency Detection51
A deep unsupervised RGB-D saliency detection approach, which requires no manual pixel-level annotation during training, and can be adapted to work in fully-supervised situation.
- DefMamba: Deformable Visual State Space Model46
This work proposes a novel visual foundation model called Def-Mamba, which includes a multi-scale backbone structure and deformable mamba (DM) blocks, which dynamically adjust the scanning path to prioritize important information, thus enhancing the capture and processing of relevant input features.
- Learning Generalized Medical Image Segmentation from Decoupled Feature Queries46
The proposed decoupled fea- ture query (DFQ) scheme can be seamlessly integrate into the Transformer segmentation model in an end-to-end manner and notably outperforming the runner-up by 1.31% and 1.98% with DSC metric on generalized fundus and prostate benchmarks, respectively.
- 44
- TW-GAN: Topology and width aware GAN for retinal artery/vein classification44
A novel Topology and Width Aware Generative Adversarial Network is proposed, which, for the first time, integrates the topology connectivity and vessel width information into the deep learning framework for A/V classification.
- 42
- 41
- Delving into Calibrated Depth for Accurate RGB-D Salient Object Detection39
A novel framework that contains a learning strategy to calibrate the latent bias in the original depth maps towards boosting the SOD performance and a boundary-aware multimodal fusion module to fuse the complementary cues from RGB and depth channels, as well as to improve object boundary qualities is proposed.
- SemanticRT: A Large-Scale Dataset and Method for Robust Semantic Segmentation in Multispectral Images36
A SemanticRT dataset is introduced - the largest MSS dataset to date, comprising 11,371 high-quality, pixel-level annotated RGB-thermal image pairs, which is 7 times larger than the existing MFNet dataset.
- 35
- 31
- Upconverting and persistent luminescent nanocarriers for accurately imaging-guided photothermal therapy27
The experimental results showed that the UPLNs@mSiO2 nanoparticles could significantly reduce the short-lived autofluorescence interference and improve signal-to-noise ratio during FPTT and may be a promising tool for improving the accuracy of PTT in vivo.
- 18
- Highly Efficient 3D Human Pose Tracking From Events With Spiking Spatiotemporal Transformer17
This work introduces the first sparse Spiking Neural Networks (SNNs) framework for 3D human pose tracking based solely on events, and eliminates the need to convert sparse data to dense formats or incorporate additional images, thereby fully exploiting the innate sparsity of input events.
- Low-complexity signal detection networks based on Gauss-Seidel iterative method for massive MIMO systems17
Improved BGS-Net, a model-driven deep learning detector network, based on the Gauss-Seidel iterative method, has lower complexity and similar the detection performance; good robustness, and its performance is less affected by changes in the number of antennas; Improved B GS-Net can improve the detection Performance.
- Local learning-based feature weighting with privacy preservation16
A local learning-based feature weighting framework is introduced and the experimental results show that the algorithms can preserve the data privacy to some extent and the objective perturbations always obtains higher classification performance than output perturbation when the privacy preserving degree is constant.
- 14
- 14
- Characteristics analysis of near-field and far-field aerodynamic noise around high-speed railway bridge13
The results show that there is the largest sound source energy on the nose tip of the high-speed train; the fluctuating pressure between air and train surface contributes the most to aerodynamic noise; the train speed, train type, and the height variation of the bridge pier can affect the distribution of far-field aerodynamics noise.
- Quantum Bacterial Foraging Optimization: From Theory to MIMO System Designs13
Improve some critical aspects such as reproduction and dispersal processes of AQBFO, propose an improved IQBFO algorithm, and apply it for interference coordination in 3D multi-cell multi-user MIMO systems, aiming to maximize the spectral efficiency.
- 13
- Cross-Level Multi-Instance Distillation for Self-Supervised Fine-Grained Visual Categorization12
The key idea is to consider the importance of each image patch in determining the fine-grained representation by multiple instance learning, which outperforms the contemporary methods by up to 10.14% and existing state-of-the-art self-supervised learning approaches by up to 19.78% on both top-1 accuracy and Rank-1 retrieval metric.
- 11
- What Limits Virtual Agent Application? OmniBench: A Scalable Multi-Dimensional Benchmark for Essential Virtual Agent Capabilities10
OmniBench is introduced, a self-generating, cross-platform, graph-based benchmark with an automated pipeline for synthesizing tasks of controllable complexity through subtask composition and OmniEval, a multidimensional evaluation framework that includes subtask-level evaluation, graph-based metrics, and comprehensive tests across 10 capabilities.
- 9
- Unleashing Multispectral Video's Potential in Semantic Segmentation: A Semi-supervised Viewpoint and New UAV-View Benchmark9
The introduction of MVUAV, a new MVSS benchmark dataset, and the development of a dedicated semi-supervised MVSS baseline - SemiMV are presented, proposed to enforce consistency regularization through a dedicated Cross-collaborative Consistency Learning (C3L) module and a denoised temporal aggregation strategy.
- 9
- Energy-based feature ranking for assessing the dysphonia measurements in Parkinson detection9
In this study, motivated by statistical learning theory, the problem of PD detection is addressed to classify the participant as healthy or PD using support vector machine (SVM) with the dysphonia measurements as the input feature vector with an energy-based feature-ranking algorithm adopted.
- 8
- 8
- 8
- 7
- 7
- 7
- 5
- 4
- 4
- 4
- 4
- 4
- 4
- 3
- 3
- 3
- 3
- Min-Max Ensemble Feature Selection3
A Min-Max Ensemble Feature Selection (M2-EFS) is proposed, which is based on balanced data partition and min-max ensemble strategy, and the experimental results demonstrate that the M2-EFS can obtain higher performance than other classical ensemble methods in most cases, especially for large scale, high dimension and imbalanced data.
- Date Fusion Schemes Based on Cooperative Spectrum Sensing3
Five fusion methods based on decision fusion and data fusion are investigated and they are used to sense spectrum cooperatively in CR networks and their spectrum detection performances are compared.
- 2
- AtlantisGS: Underwater Sparse-View Scene Reconstruction via Gaussian Splatting2
AtlantisGS incorporates a scattering decomposition method that separates medium and object contributions during rendering, and a sparse Gaussian proliferation strategy that adaptively densifies the scene representation to improve structural accuracy.
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- Energy-efficient cooperative routing protocol for wireless sensor networks2
A novel cooperative routing protocol is proposed in wireless sensor networks and a minimum energy multi-nodes cooperative path is constructed by the cooperative transmission of neighboring nodes and comparison of total power consumption.
- 1
- MetaComp: Learning to Adapt for Online Depth Completion1
Experimental results and comprehensive ablations show that the proposed MetaComp is capable of adapting to the depth completion in a new environment effectively and robust to changes in different modalities.
- 1
- 1
- 1
- 1
- Spike Sorting with Locally Weighted Co-association Matrix-Based Spectral Clustering1
A locally weighted co-association matrix is employed as the similarity matrix and the Shannon entropy is introduced to measure the dependability of clustering to improve the performance of spectral clustering algorithm for spike sorting.
- 1
- 1
- Three-Dimension Kronecker Channel Modeling and Correlation Analysis1
A three-dimension 3D multiple-input multiple-output MIMO channel model is derived by considering the elevation dimension and the azimuth dimension together, which is developed for arbitrary antenna arrays with non-isotropic antenna patterns and also for any propagation environment of 3D MIMo systems.
- Local energy-based framework for feature ranking1
A local energy-based framework is proposed to estimate the features' relevance for ranking them and its performance is higher or similar to other classical feature ranking algorithms in most cases.
- Discriminant analysis and prediction of house prices based on support vector machines1
The result shows that SVM has certain advantages for the research of small sample in house prices and two kinds of regression analysis in multivariate statistical analysis.
- Enhanced cooperative packet retransmission employing joint cooperative diversity and symbol mapping diversity1
A joint cooperation diversity and symbol mapping diversity (SMD) strategy is proposed for cooperative packet retransmission system with high-order modulation such as 16QAM and simulation results show that the joint-diversity strategy can provide more BER gains than other relaying strategies.
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- A normative theory of visual working memory limitations.–
This work conceptualizes VWM as a system that balances between two conflicting goals: keeping memory errors small and spiking activity low, and formalizes this trade-off in a loss function and shows that minimization of loss dictates a strategy in which memory precision declines with the number of remembered items.
- –
- –
- –
- –
- Design and Implementation of a Micro-Grid Monitoring System–
This paper designs a practical micro-grid system including solar power, wind power and energy storage, and the monitoring system is tested on a practical project, which proves the feasibility of the software and the design.
- –
- –
- The Review of Debate on the Classification of Knowledge Mode 1 and Mode 2 in the UK Higher Education–
- Probe into Improving the Quality of University Library's Great Circulation Service through Uniforming–
This paper probes into the functions of uniforming in improving university library's great circulation service, and puts forward some principles of uniformed service.
- A Cooperative ARQ Protocol Based on Conditional Probability Retransmission–
In cognitive radio networks with multiple second users, the total second networks throughput increases if the system uses a new C-ARQ protocol based on conditional probability retransmission and implements acknowledgment signals from the relays.
- –
- –
- –
- Impact of CET-IBT on College English Teaching–
CET-IBT is an important method of testing listening speaking ability and comprehensive level of College English, with the advantages of perfect systematization, high efficiency and saving cost etc.
- Topology-Transparent Hybrid MAC Protocol for Ad Hoc Networks–
Simulation results show that the proposed TTHM protocol is better than the T-TSMA protocol and suitable for distributed application since it keeps the advantage of the topology transparency and eliminates the maximum nodal degree constraint.
- –
- The Effective Simulation of Wireless Ad Hoc Network under Fading Channel–
A new bit based signal reception model is proposed that shows that by the proposed bit based model the frame error rate(FER)increases with the velocity while by the frame based modelThe FER is independent of velocity.
- A New MPR Algorithm for Ad hoc Networks–
This new method improves the correct ratio of recovered packets by using decision feedback and channel estimation schemes to reduce the bit error ratio of the recovered packets and improves the performance of the MPR in Ad hoc Networks.
- Performance Analysis of 802.11 DCF over Mobile Ad Hoc Networks–
The study shows that the throughput of 802.11 DCF over MANET correlated fading channel decreases with the increase of mobile velocity and there exists an optimal frame length to maximize throughput under a certain velocity.
- Comment on the cognitive theory of acquisition of second language lexicon–
The cognitive theory illustrates the mechanism of L2 lexicon acquisition from the cognitive subject and object and comment are made from monitor theory, schema theory and interlanguage theory to provide the reference for the study of L1 teaching and lexicon Acquisition.
- Improving 2D Virtual ESPRIT Algorithm–
Modifications of the 2D virtual ESPRIT algorithm are made that make better use of the sensors, or relaxed the requirements of the sensor array at the similar computational cost.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. 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.