Research Profile
Research Experience
3D Perception and Reconstruction from Sparse Multi-view Observations
Research on multi-view observation modelling, image enhancement, and learning-based 3D reconstruction under sparse, noisy, and complex observation conditions.
- Multi-view Geometry: Modelled geometric relationships between multi-view observations and 3D target structures for sparse-view reconstruction.
- Multi-view Fusion: Investigated cross-view information fusion to improve reconstruction robustness from limited observations.
- Learning-Based Reconstruction: Explored NeRF- and 3DGS-based methods for high-fidelity 3D reconstruction from sparse multi-view observations.
- Image Enhancement: Developed self-supervised methods for image denoising and structure preservation without requiring clean reference data.
- Physics-Based Reconstruction: Developed reconstruction methods for low-SNR and complex-motion observations.
UAV-Based Sensing and 3D Reconstruction
Developed UAV-based sensing, image processing, and multi-view 3D reconstruction methods for complex urban environments.
- Multi-view Data Acquisition: Conducted 40+ UAV sorties at 170–260 m for real-world sensing and data collection.
- Image Processing: Developed physics-based motion estimation and self-supervised denoising methods for low-SNR and complex-motion observations, improving SNR by 10–15 dB.
- 3D Reconstruction: Developed multi-view reconstruction methods using ADMM-Net and 3DGS, achieving sub-meter reconstruction accuracy for representative urban scenes.
- Real-World Validation: Conducted 100+ experiments on TB-scale real sensor data collected in complex environments.
mmWave Radar-Based Human Activity Recognition
Developed learning-based human gesture recognition methods covering sensor data acquisition, signal processing, feature extraction, feature fusion, and classification.
- Data Acquisition: Built an AWR1642 mmWave radar sensing system and collected data from 10 gesture classes.
- Signal Processing: Applied filtering and CFAR detection to extract temporal range, velocity, and angle features.
- Feature Fusion: Developed a dual-branch network to integrate complementary motion features for dynamic gesture classification.
- Results: Achieved >94.5% recognition accuracy through data augmentation and model optimization.
Selected Publications & Patents
A Parametric 3-D ISAR Imaging Method of Celestial Target Under Low SNR
IEEE Transactions on Geoscience and Remote Sensing | Co-author
Published
Multi-Dimensional Spread Target Detection with Across Range-Doppler Unit Phenomenon Based on Generalized Radon-Fourier Transform
Remote Sensing | First Author
Published
An Adaptive 3-D Reconstruction Method for Targets Based on Multi-view Self-supervised Framework under Low SNR
IEEE TAES | First Author
Under Review
Scattering-Aware Multi-View Masked Networks for Self-Supervised Radar Denoising
IEEE Transactions on Geoscience and Remote Sensing | First Author
Under Review
A Self-supervised Radar Sparse Imaging Method via Physics-Aware Imputation Network
IEEE Transactions on Aerospace and Electronic Systems | First Author
Under Review
Method for Multi-view 3D Reconstruction under Low SNR
Chinese Invention Patent | First Student Inventor
Granted
Image Denoising Method Based on Multi-angle Observations and Self-supervised Learning
Chinese Invention Patent | First Student Inventor
Granted
Self-Supervised Image Denoising Method Based on an Adaptive Masking Strategy
Chinese Invention Patent | First Student Inventor
Patent Application
Image Reconstruction Method Based on a Self-Supervised Inpainting Network
Chinese Invention Patent | First Student Inventor
Patent ApplicationEducation
Relevant Coursework: Signals and Systems, Digital Signal Processing, Communication Principles.
Technical Skills
-
3D Computer Vision
- Multi-view geometry
- Multi-view reconstruction
- Sparse-view reconstruction
- 3D perception
-
Neural 3D Reconstruction
- NeRF
- 3D Gaussian Splatting (3DGS)
- Deep unfolding / ADMM-Net
-
Deep Learning
- Self-supervised learning
- Transformer
- Diffusion models
- CNN-based models
-
Image Processing
- Image enhancement and denoising
- Sparse image reconstruction
- Structure-preserving reconstruction
-
Programming
- Python / PyTorch
- MATLAB
- C / C++
-
Experimental Research
- UAV-based data acquisition
- Real-world sensor experiments
- Multi-source data processing
- Algorithm validation and performance evaluation
-
3D & Simulation Tools
- MeshLab
- COLMAP
- CST Studio / FEKO
- STK
-
Sensor Platforms
- mmWave radar
- LiDAR
- UAV sensing systems
- Anechoic chamber experiments
Honors & Awards
China Scholarship Council Scholarship
Funded joint PhD research at the University of Auckland.
Beijing Outstanding Graduate
Recognized for outstanding academic achievement and comprehensive performance.
First-Class Scholarships
Received multiple municipal- and university-level scholarships for academic excellence.
National Level-II Athlete Standard in Marathon Running
Long-term endurance athlete with 20+ races completed.
AHA / Red Cross First Aid Instructor
Certified first aid instructor with experience supporting large-scale events.
Outstanding Student Leader (3 Awards)
Recognized three times for leadership, teamwork, and contributions to student activities.
Leadership & Activities
Summer Teaching Volunteer Program, China
- Initiated and organized educational outreach programs in rural areas, coordinating volunteer recruitment, curriculum design, school engagement, and team management.
American Heart Association & Beijing Red Cross
- Delivered CPR and first-aid training to more than 1,000 participants across universities, companies, and public events.
