Area of research
Computer Vision and Pattern Recognition
Research interest
Research interests include Advanced Image and Video Retrieval Techniques, Video Analysis and Summarization, Image Retrieval and Classification Techniques, and Human Pose and Action Recognition.
Radiomics-Guided Deep Learning-Based Framework for Few-Shot Automatic CTV Delineation of Endometrial Carcinoma in Multi-Center
GAT-COBO: Cost-Sensitive Graph Neural Network for Telecom Fraud Detection
Exploring Fine-Grained Representation and Recomposition for Cloth-Changing Person Re-Identification
FastOcc: Accelerating 3D Occupancy Prediction by Fusing the 2D Bird’s-Eye View and Perspective View
VPL-SLAM: A Vertical Line Supported Point Line Monocular SLAM System
Exploring One-Shot Semi-supervised Federated Learning with Pre-trained Diffusion Models
Simulation and assimilation of the digital human brain
Generating and Reweighting Dense Contrastive Patterns for Unsupervised Anomaly Detection
Grad-PU: Arbitrary-Scale Point Cloud Upsampling via Gradient Descent with Learned Distance Functions
Cost-Sensitive GNN-Based Imbalanced Learning for Mobile Social Network Fraud Detection
Chinese Text Recognition with A Pre-Trained CLIP-Like Model Through Image-IDS Aligning
Co-attention Aligned Mutual Cross-Attention for Cloth-Changing Person Re-identification
Metaverse: Perspectives from graphics, interactions and visualization
SAR-Net: Shape Alignment and Recovery Network for Category-level 6D Object Pose and Size Estimation
Density-preserving Deep Point Cloud Compression
SGM3D: Stereo Guided Monocular 3D Object Detection
RCLane: Relay Chain Prediction for Lane Detection
Long-Term Cloth-Changing Person Re-identification
Scene Text Telescope: Text-Focused Scene Image Super-Resolution
Depth-conditioned Dynamic Message Propagation for Monocular 3D Object Detection
Learning Dynamic Alignment via Meta-filter for Few-shot Learning
Zero-Shot Chinese Character Recognition with Stroke-Level Decomposition
Temporal Context Aggregation for Video Retrieval with Contrastive Learning
Delving into Data: Effectively Substitute Training for Black-box Attack
The Devil is in the Task: Exploiting Reciprocal Appearance-Localization Features for Monocular 3D Object Detection
BERT-ATTACK: Adversarial Attack Against BERT Using BERT
DeepSFM: Structure from Motion via Deep Bundle Adjustment
Pixel2Mesh: 3D Mesh Model Generation via Image Guided Deformation
Deep learning for detecting corona virus disease 2019 (COVID-19) on high-resolution computed tomography: a pilot study
M$^3$Lung-Sys: A Deep Learning System for Multi-Class Lung Pneumonia Screening From CT Imaging