Hanwang Zhang lists you as their PhD student. Claim this profile to confirm it and keep the rest of your record right.
Claim this profileAcademic lineage
View as a treeAdvisors
Works132 from public data
- Unbiased Scene Graph Generation From Biased Training874
A novel SGG framework based on causal inference but not the conventional likelihood is presented, which uses Total Direct Effect (TDE) as the proposed final predicate score for unbiased SGG and can be widely applied in the community who seeks unbiased predictions.
- 673
- NPA374
A neural news recommendation model with personalized attention (NPA) which exploits the embedding of user ID to generate the query vector for the word- and news-level attentions.
- Improving 3D Object Detection with Channel-wise Transformer316
This paper leverages the high-quality region proposal network and a Channel-wise Transformer architecture to constitute the two-stage 3D object detection framework (CT3D) with minimal hand-crafted design and achieves superior performance and excellent scalability.
- Towards Federated Learning in UAV-Enabled Internet of Vehicles: A Multi-Dimensional Contract-Matching Approach313
This work proposes the adoption of a Federated Learning (FL) based approach to enable privacy-preserving collaborative Machine Learning across a federation of independent DaaS providers for the development of IoV applications, e.g., for traffic prediction and car park occupancy management.
- Visual Commonsense R-CNN299
A novel unsupervised feature representation learning method, Visual Commonsense Region-based Convolutional Neural Network (VC R-CNN), is presented to serve as an improved visual region encoder for high-level tasks such as captioning and VQA, and observes consistent performance boosts across them.
- A Survey on Deep Hashing Methods278
This survey detailedly investigates current deep hashing algorithms including deep supervised hashing and deep unsupervised hashing, and categorizes deep supervised hash methods into pairwise methods, ranking-based methods, pointwise methods as well as quantization according to how measuring the similarities of the learned hash codes.
- Camera-Aware Proxies for Unsupervised Person Re-Identification268
This paper proposes to split each single cluster into multiple proxies and each proxy represents the instances coming from the same camera, and designs both intra and inter-camera contrastive learning components for the Re-ID model to effectively learn the ID discrimination ability within and across cameras.
- Gradient Centralization: A New Optimization Technique for Deep Neural Networks228
It is shown that GC can regularize both the weight space and output feature space so that it can boost the generalization performance of DNNs, and improves the Lipschitzness of the loss function and its gradient so that the training process becomes more efficient and stable.
- Cloth-Changing Person Re-identification from A Single Image with Gait Prediction and Regularization200
This paper introduces Gait recognition as an auxiliary task to drive the Image ReID model to learn cloth-agnostic representations by leveraging personal unique and cloth-independent gait information, and introduces this framework as GI-ReID, which enables an efficient and latency-free person identity matching for surveillance.
- Graph Contrastive Clustering175
This work first proposes a novel graph contrastive learning framework, and then applies it to the clustering task, resulting in the Graph Constrastive Clustering (GCC) method, which incorporates the latent category information to reduce the intra-clusters variance as well as increase the inter-cluster variance.
- Two Causal Principles for Improving Visual Dialog164
This paper unravels the design tricks adopted by the champion team MReaL-BDAI for Visual Dialog Challenge 2019: two causal principles for improving VisDial, which are model-agnostic, so they are applicable in any VisDial model.
- Elevation of blood glucose level predicts worse outcomes in hospitalized patients with COVID-19: a retrospective cohort study162
The findings may provide a simple and practical way to risk stratify COVID-19 inpatients for hierarchical management, particularly where medical resources are in severe shortage during the pandemic.
- Joint Auction-Coalition Formation Framework for Communication-Efficient Federated Learning in UAV-Enabled Internet of Vehicles158
A joint auction-coalition formation framework to solve the allocation of UAV coalitions to groups of IoV components and it is shown that as the cooperation cost of the Uavs increases, the UAVs prefer to support the Iov components independently and not to form any coalition.
- 157
- MaCAR: Urban Traffic Light Control via Active Multi-agent Communication and Action Rectification126
This work proposes a novel Multi-agent Communication and Action Rectification (MaCAR) framework which enables active communication between agents by considering the impact of synchronous actions of agents and outperforms state-of-the-art methods on both synthetic and real-world datasets.
- Online Convolutional Reparameterization101
This paper presents online convolutional re-parameterization (OREPA), a two-stage pipeline aiming to reduce the huge training overhead by squeezing the complex training-time block into a single convolution, and introduces a linear scaling layer for better optimizing the online blocks.
- Structural and Statistical Texture Knowledge Distillation for Semantic Segmentation92
This paper introduces a Contourlet Decomposition Module (CDM) that decomposes low-level features with iterative laplacian pyramid and directional filter bank to mine the structural texture knowledge and proposes a Denoised Texture Intensity Equalization Module (DTIEM) to adaptively extract and enhance statistical texture knowledge through heuristics iterative quantization and denoised operation.
- DCT-Mask: Discrete Cosine Transform Mask Representation for Instance Segmentation85
This paper proposes a new mask representation by applying the discrete cosine transform(DCT) to encode the high-resolution binary grid mask into a compact vector, termed DCT-Mask, which could be easily integrated into most pixel-based instance segmentation methods.
- Spatio-Temporal Inception Graph Convolutional Networks for Skeleton-Based Action Recognition81
This work designs a simple and highly modularized graph convolutional network architecture for skeleton-based action recognition by repeating a building block that aggregates multi-granularity information from both the spatial and temporal paths.
- Dynamic Spatio-Temporal Graph-Based CNNs for Traffic Flow Prediction80
DST-GCNN is a two stream network that learns expressive features to represent spatio-temporal structures and predicts future traffic flows from surveillance video data and achieves competitive performances compared with the other state-of-the-art methods.
- Dense Interaction Learning for Video-based Person Re-identification77
A hybrid framework that takes the principal advantages of both CNN-based and Attention-based architectures to tackle video-based person re-ID difficulties and introduces Spatio-TEmporal Positional Embedding into the DI decoder to investigate the positional relation among the spatial-temporal inputs.
- 74
- Traffic Flow Prediction with Vehicle Trajectories68
A spatiotemporal deep learning framework that mines the underlying causality of flows from historical vehicle trajectories and incorporates that into road traffic prediction and achieves over 5% error reduction when compared with the state-of-the-art approaches.
- Systemic Corticosteroids and Mortality in Severe and Critical COVID-19 Patients in Wuhan, China68
Absence of the beneficial effect in this study in contrast to that observed in the RECOVERY clinical trial may be due to biases in observational data, in particular prescription by indication bias.
- 68
- 66
- Large-scale vehicle trajectory reconstruction with camera sensing network58
VeTrac, a comprehensive system that employs widely deployed traffic cameras as a sensing network to trace vehicle movements and reconstruct their trajectories in a large scale, fuses mobility correlation and vision-based analysis to reduce uncertainties in identifying vehicles.
- 54
- City brain: practice of large‐scale artificial intelligence in the real world54
The authors introduce current practices of the City Brain platform in a few cities in China, including what they can do to achieve the goal and make it a reality, and focus on the system overview and key technical details of each component of the city Brain system, from cognition to intervention.
- Local Convolutional Neural Networks for Person Re-Identification53
For the task of person re-identification, the local convolutional neural networks (Local CNN) can outperform state-of-the-art methods consistently on three large-scale benchmarks, including Market-1501, CUHK03, and DukeMTMC-ReID.
- The Blessings of Unlabeled Background in Untrimmed Videos50
A Temporal Smoothing PCA-based (TS-PCA) deconfounder is proposed, which exploits the unlabelled background to model an observed substitute for the unobserved confounder, to remove the confounding effect of weakly-supervised Temporal Action Localization.
- 50
- Homography Loss for Monocular 3D Object Detection48
A differentiable loss function, termed as Homography Loss, is proposed to achieve the goal, which exploits both 2D and 3D information, aiming at balancing the positional relationships between different objects by global constraints, so as to obtain more ac-curately predicted 3D boxes.
- 48
- Category Dictionary Guided Unsupervised Domain Adaptation for Object Detection48
A category dictionary guided (CDG) UDA model for cross-domain object detection is proposed, which learns category-specific dictionaries from the source domain to represent the candidate boxes in target domain and a residual weighted self-training paradigm is developed to implicitly align source and target domains for detection model training.
- 47
- Apparel-Invariant Feature Learning for Person Re-Identification45
The semi-supervised Apparel-invariant Feature Learning (AIFL) framework is proposed to learn an apparel- Invariant pedestrian representation using images of the same person wearing different clothes, and an unsupervised apparel-simulation GAN (AS-GAN) to synthesize cloth-changing images according to the target cloth embedding.
- CIMON: Towards High-quality Hash Codes44
A new method named Comprehensive sImilarity Mining and cOnsistency learNing (CIMON) is proposed, which uses global refinement and similarity statistical distribution to obtain reliable and smooth guidance and both semantic and contrastive consistency learning are introduced.
- Deconfounded Visual Grounding40
This work frames the visual grounding pipeline into a causal graph, which shows the causalities among image, query, target location and underlying confounder, and proposes a confounders-agnostic approach called Referring Expression Deconfounder (RED), to remove the confounding bias.
- STP-TrellisNets: Spatial-Temporal Parallel TrellisNets for Metro Station Passenger Flow Prediction39
A novel deep learning framework STP-TrellisNets is proposed, which for the first time augments the newly-emerged temporal convolutional framework TrellisNet for spatial-temporal prediction and outperforms the state-of-the-art baselines.
- 38
- Concentrated Local Part Discovery With Fine-Grained Part Representation for Person Re-Identification36
- Diagnostic value of platelet indexes for pulmonary embolism36
The combined application of MPV can improve the diagnostic value of d-dimer for PE and increase the area under the curve in patients with PE.
- A Statistical Approach to Mining Semantic Similarity for Deep Unsupervised Hashing35
A general distribution-based metric is proposed to depict the pairwise distance between images and is applied to deep unsupervised hashing, resulting in Distribution-based similArity sTructure rEconstruction (DATE), which can generate more accurate semantic similarity information by using non-parametric ball divergence.
- 35
- Computer-Aided Diagnosis of Laryngeal Cancer Based on Deep Learning with Laryngoscopic Images34
This study demonstrates that Densenet201 is a highly accurate and reliable tool for the computer-aided diagnosis of laryngeal cancer based on laryngoscopic images and underscores the potential of deep learning as a complementary tool for clinicians and the importance of incorporating advanced technology in improving diagnostic accuracy and patient care in laryngeal cancer diagnosis.
- 34
- 32
- 31
- 31
- Intuitive and efficient roof modeling for reconstruction and synthesis30
A novel and flexible roof modeling approach that can be used for constructing planar 3D polygon roof meshes by using transformers and graph convolutional networks to model roof topology and enforces the roof validity by optimizing a simple but effective planarity metric.
- Deep Position-wise Interaction Network for CTR Prediction30
A Deep Position-wise Interaction Network (DPIN) is proposed to efficiently combine all candidate items and positions for estimating CTR at each position, achieving consistency between offline and online as well as modeling the deep non-linear interaction among position, user, context and item under the limit of serving performance.
- 29
- 29
- 27
- Towards Precise Intra-camera Supervised Person Re-Identification26
This paper proposes jointly learned camera-specific non-parametric classifiers, together with a hybrid mining quintuplet loss, to perform intra-camera learning and shows that this approach outperforms all existing ICS works by a great margin.
- 24
- 21
- Based on CT at the third lumbar spine level, the skeletal muscle index and psoas muscle index can predict osteoporosis20
This study demonstrated that L3-based muscle index could assist clinicians in the diagnosis of osteoporosis to a certain extent, and PMI is superior to SMI in thediagnosis of osteopsorosis.
- 19
- Deep Adaptive Graph Clustering via von Mises-Fisher Distributions18
Deep Adaptive Graph Clustering via von Mises-Fisher distributions is proposed, namely DAGC, which can simultaneously improve the intra-cluster compactness and inter-cluster heterogeneity.
- Study on risk factors of preoperative deep vein thrombosis in patients with lower limb fractures and construction and validation of risk prediction nomogram model16
Early monitoring of D-D, FIB, FDP and PLT levels in patients with lower limb fracture can effectively screen for lower limb deep vein thrombosis.
- Meta Clustering Learning for Large-scale Unsupervised Person Re-identification16
This paper makes the first attempt to the large-scale U-ReID and proposes a "small data for big task" paradigm dubbed Meta Clustering Learning (MCL), which only pseudo-labels a subset of the entire unlabeled data via clustering to save computing for the first-phase training.
- Dual Graph for Traffic Forecasting15
A novel dual graph framework, called DualGraph, is proposed, to model the propagation behavior of traffic on road networks, and shows that even for node or edge traffic forecasting alone, the model still outperforms compared ones, especially for long term prediction.
- Homocentric Hypersphere Feature Embedding for Person Re-Identification15
A novel homocentric hypersphere embedding scheme is proposed to decouple magnitude and orientation information for both feature and weight vectors, and reformulate the triplet loss and the softmax loss to their angular versions and combine them into an angular discriminative loss.
- Networked Time-series Prediction with Incomplete Data via Generative Adversarial Network14
This article proposes networked time series Imputation Generative Adversarial Network (NETS-ImpGAN), a novel deep learning framework that can be trained on incomplete data with missing values in both history and future, and proposes Graph Temporal Attention Networks, which incorporate the attention mechanism to capture both inter-time series and temporal correlations.
- 14
- 14
- Criterion-based Heterogeneous Collaborative Filtering for Multi-behavior Implicit Recommendation13
A novel non-sampling learning framework named Criterion-guided Heterogeneous Collaborative Filtering (CHCF), which integrates criterion learning and user preference learning into a unified framework, which can be trained jointly for the interaction prediction of the target behavior.
- Screening of linear B-cell epitopes and its proinflammatory activities of Haemophilus parasuis outer membrane protein P213
A linear B-cell epitope map of the OmpP2 protein is profiled and preliminarily analyzed the proinflammatory activities and effects of these epitopes on bacterial virulence, providing a reliable theoretical basis for establishing a method to distinguish strain pathogenicity and to screen candidate peptides for subunit vaccines.
- A topography-aware approach to the automatic generation of urban road networks13
A new method of street network generation based on a generative adversarial network by designing a pre-positioned geo-extractor module and a geo-merging bypath is proposed, which yields a more realistic prediction of street configurations than conventional image inpainting techniques.
- Stylized Adversarial AutoEncoder for Image Generation13
An autoencoder-based generative adversarial network (GAN) for automatic image generation, which is called "stylized adversarial autoenCoder", which achieves superior results for image generation as well as remarkably improves the corresponding supervised recognition task.
- 12
- 12
- Graph-Induced Contrastive Learning for Intra-Camera Supervised Person Re-Identification11
This article first formulate the cross-camera ID association task as a graph partitioning problem subjected to ICS-specific constraints and design a greedy agglomeration algorithm to solve it, and proposes a graph-induced contrastive loss that unifies both intra- and inter-camera learning into a contrastive learning framework to learn a Re-ID model.
- Progressive Transfer Learning for Person Re-identification11
This paper introduces a novel network structure called Batch-related Convolutional Cell (BConv-Cell), which progressively collects the global information of the dataset into a latent state and uses this latent state to rectify the extracted feature.
- Global leaders malnutrition initiative-defined malnutrition affects long-term survival of different subgroups of patients with gastric cancer: A propensity score-matched analysis10
Subgroup analysis showed that the GLIM was more appropriate for predicting OS in older aged patients (≥65 years), females, those with comorbidities, and those with advanced gastric cancer (TNM stage = 3) than the other way around.
- Learning to teach and learn for semi-supervised few-shot image classification10
The proposed LTTL combines the power of meta-learning and self-training, achieving superior performance compared with the baseline methods on two public benchmarks, and uses the meta- learning paradigm to optimize the parameters in the whole framework.
- AutoSmart: An Efficient and Automatic Machine Learning Framework for Temporal Relational Data10
The proposed framework, AutoSmart, is the winning solution to the KDD Cup 2019 of the AutoML Track, which is one of the largest AutoML competition to date and outperforms the baseline solution significantly on several datasets in various domains.
- FGAGT: Flow-Guided Adaptive Graph Tracking.9
This article proposes the FGAGT tracker, which reaches the level of state-of-the-art, where the MOTA index exceeds FairMOT by 2.5 points, and CenterTrack by 8.4 points on the MOT17 dataset.
- ST4ML: Machine Learning Oriented Spatio-Temporal Data Processing at Scale8
ST4ML is presented, a distributed spatio-temporal data processing system to support scalable machine-learning-oriented applications based on Apache Spark and is the first of its kind to realize design considerations.
- Pursuing Knowledge Consistency: Supervised Hierarchical Contrastive Learning for Facial Action Unit Recognition8
A supervised hierarchical contrastive learning method (SupHCL) is proposed for AU recognition to pursue knowledge consistency among different facial images and different AUs, which is orthogonal to methods focusing on network architecture design.
- Neural CTR Prediction for Native Ad8
A neural approach for native ad CTR prediction is proposed which can incorporate different kinds of user behaviors to model user interests, and can fully exploit the textual information in ads anduser behaviors to learn accurate ad and user representations.
- Decouple co-adaptation: Classifier randomization for person re-identification7
This paper proposes a very simple and effective training method, called DeAda, to construct a series of weak classifiers during training by randomization of parameters, so that optimization on the feature extractor could be strengthened in the training stage.
- BERP: A Blind Estimator of Room Parameters for Single-Channel Noisy Speech Signals6
BERP is proposed, a new universal approach to blindly estimate RAPs, RGPs, and occupancy levels that achieves state-of-the-art performance and excellent adaptability to real-world scenarios.
- Panet: Perspective-Aware Network with Dynamic Receptive Fields and Self-Distilling Supervision for Crowd Counting6
This paper proposes a novel perspective-aware approach called PANet, which outperforms the state-of-the-art methods by a large margin and proposes the dynamic receptive fields (DRF) framework, which is able to adjust the receptive field by the dilated convolution parameters according to the input image.
- Second-Order Camera-Aware Color Transformation for Cross-Domain Person Re-identification6
A Second-order Camera-aware Color Transformation (SCCT) that can operate on image level and align the second-order statistics of all the views of both source and target domain data with original ImageNet data statistics is proposed.
- Part-Aware Attention Network for Person Re-identification6
A novel Part-aware Attention Network (PAN) is proposed to connect part feature maps and middle-level features and achieves leading performance on Re-ID benchmarks Market1501, DukeMTMC, and CUHK03.
- Spatial likelihood voting with self-knowledge distillation for weakly supervised object detection6
In this framework, a spatial likelihood voting (SLV) module is introduced to converge region proposal localization without bounding box annotations and a self-knowledge distillation module is proposed to refine the feature representations of the given image.
- Effect of far-infrared therapy device on arteriovenous fistula maturation and lifespan in hemodialysis patients: a randomized controlled clinical trial5
As a promising new treatment method, far infrared therapy can effectively promote the maturity of AVF, increase venous diameter, pump controlled blood flow during dialysis, brachial artery blood flow in the intervention group, and prolong the service life of AVf.
- Analysis of safety and efficacy of conservative treatment and endovascular treatment in patients with spontaneous isolated mesenteric artery dissection5
Conservative treatment is effective for most symptomatic SISMAD patients, and close monitoring is required; for patients with persistent symptoms and severe true lumen stenosis (especially Yun classification type III), endovascular treatment is preferred.
- 5
- A Comprehensive Benchmark for Electrocardiogram Time-Series4
This paper presents an in-depth investigation of ECG signals and establishes a comprehensive benchmark, which includes identifying limitations in traditional evaluation metrics for ECG analysis, and introducing a novel metric and model architecture.
- 4
- 3
- 3
- Attention-based Class Activation Diffusion for Weakly-Supervised Semantic Segmentation3
A new method to couple CAM and Attention matrix in a probabilistic Diffusion way is proposed, and it is shown that AD-CAM as pseudo labels can yield stronger WSSS models than the state-of-the-art variants of CAM.
- Research on improvement of Louvain community detection algorithm3
This paper proposes the following improvements for Louvain algorithm: introduce WCC algorithm, so thatLouvain can be applied to disconnected graphs and has better partitioning effect and faster processing speed.
- A Testbed for Studying COVID-19 Spreading in Ride-Sharing Systems3
A multi-agent testbed to study the spread of infectious diseases through ride-sharing systems that allows users to vary the parameters of the disease and behaviours toStudy the interaction effect between technology, disease and people's behaviours in such a complex environment.
- 3
- 3
- Experimental insights into genome reconstruction driven by horizontal transfer of supernumerary chromosomes in Magnaporthe oryzae2
It is found that SCs could be horizontally transferred between M. oryzae strains and revealed frequent structural variations facilitated by SCs, including deletions, duplications, translocations, and SC‐core chromosome recombinations during and after horizontal transfer.
- A multivariate time series data classification method based on dual-branch structure GSTT2
Experimental evaluations on 30 datasets from the UEA multivariate time series benchmark demonstrate that the GSTT significantly outperforms baseline methods in classification accuracy, demonstrating the superior performance of GSTT in a supercomputing environment.
- The effects of propranolol on the biology and Notch signaling pathway of human umbilical vein endothelial cells2
Propranolol may play a role in IH treatment by increasing Jagged1 expression in endothelial cells, activating the Notch pathway and inducing the upregulation of the downstream target gene HEY1.
- Multivariate deep learning for reconstruction of spatial missing climate data2
A new deep learning method is proposed that combines other variables related to missing meteorological variables to provide more effective information for missing regions and proves the good effect of the method through a series of experiments.
- 2
- Dynamic supervisor for cross-dataset object detection2
It is pointed out that it is not enough to generate high-quality annotations using a single model, which only looks once for annotations, and a dynamic supervisor framework that updates the annotations multiple times through multiple-updated submodels trained using hard and soft labels is proposed.
- Spatiotemporal Multi-Task Network for Human Activity Understanding2
A spatiotemporal, multi-task, 3D deep convolutional neural network to detect actions in untrimmed videos and introduces a novel video representation, interlaced images, as an additional network input stream to better utilize the rich motion information in videos.
- 2
- The Xenopyricularia zizaniicola exhibits a genome architecture distinct to the two-speed genome1
This study assembled the first high-quality genome of Xenopyricularia zizaniicola strain JB-1, which infects Z. latifolia, and revealed its distinct genome architecture, which closely resembles the “one-speed genome” structure, which contrasts with Pyricularia oryzae.
- Fine spatial-temporal density mapping with optimized approaches for many-core system1
Overall, the spatial-temporal density mapping method improves system performance by a factor of 1.85 compared to the commonly employed layer-wise mapping method, effectively balancing spatial and temporal resource utilization.
- 1
- 1
- 1
- 1
- Delving into Details: Synopsis-to-Detail Networks for Video Recognition1
The proposed synopsis-to-detail networks enable us to train models directly from scratch in an end-to-end manner and to investigate various architectures for synopsis/detail recognition.
- 1
- 1
- –
- –
- –
- –
- Qklu: a two-dimensional block-cyclic sparse direct solver–
This work adapted OpenMP’s dynamic multithreading scheduling to leverage OpenMP’s dynamic multithreading scheduling, utilizing the depend clause to manage dependencies between blocks, offering significantly greater flexibility.
- EAGLE: An Efficient Global Attention Lesion Segmentation Model for Hepatic Echinococcosis–
EAGLE is a U-shaped network composed of a Progressive Visual State Space (PVSS) encoder and a Hybrid Visual State Space (HVSS) decoder that work collaboratively to achieve efficient and accurate segmentation of hepatic echinococcosis (HE) lesions.
- Functional development of midbrain and pons during human gestation and electrophysiological markers for DS intervention–
The study presents the first functional map of midbrain and pons development in human fetuses from GW10 to GW17, revealing changes in neuronal excitability, synaptic transmission, and plasticity during this critical developmental window.
- –
- –
- –
- –
- –
- Comparing and Evaluating the two Diagnostic Criteria of Severity of COPD Adopted by China and America Based on Data From 713 COPD Patients–
The diagnostic criteria of severity of COPD adopted by ATS in 1995 was more objective and accurate to reflect the degree of injury of lung disease and pulmonary function, the ability of working, laboring and self-care of life.
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.