Welcome!
I am a Ph.D. candidate at Chang’an University, conducting research on situational awareness in nearshore environments. My supervisor is Professor Li Wei. My work focuses on multimodal fusion, multi-object tracking, and vision-language large models. Currently, I am deeply involved in AI Agent projects for specific scenarios as part of my doctoral research. In addition, I have conducted research on traditional computer vision (CV), edge computing, and C++ front-end development (QT) and have independently developed a complete set of front-end and back-end algorithms for industrial production (Digital Radiography imaging), which generated millions in value. If you’re interested in my work, feel free to email me.
🔥 News
- 2024.12: 🎉🎉 New Journal Paper on Modal-Invariant Progressive Representation for Multimodal Image Registration Accepted by Information Fusion (中科院一区, Top, IF=18.6, 第一作者).
- 2024.12: 🎉🎉 New Conference Paper on Cross-Modality Fusion Mamba for All-in-One Extreme Weather-Degraded Image Restoration Accepted by IEEE International Conference on Acoustics, Speech and Signal Processing 2025 (CCF-B, EI, 第一作者).
- 2024.12: 🎉🎉 New Journal Paper on LG-Diff: Learning to Follow Local Class-Regional Guidance for Nearshore Image Cross-modality High-Quality Translation Accepted by Information Fusion (中科院一区, Top, IF=18.6, 第一作者).
- 2024.10: 🎉🎉 New Journal Paper on SeaTrack: Rethinking Observation-Centric SORT for Robust Nearshore Multiple Object Tracking Accepted by Pattern Recognition (中科院一区, Top, IF=8.0, 第一作者).
- 2024.10: 🎉🎉 New Conference Paper on Learning to Follow Frequency View Guidance for Dental CT Images Deblurring Accepted by IEEE International Conference on Bioinformatics and Biomedicine 2024 (CCF-B, EI, 第一作者).
- 2024.10: 🎉🎉 New Journal Paper on Crack instance segmentation using splittable transformer and position coordinates Accepted by Automation in Construction (中科院一区, Top, IF=10.3, 通讯作者).
- 2024.09: 🎉🎉 New Journal Paper on Nearshore optical video object detector based on temporal branch and spatial feature enhancement Accepted by Engineering Applications of Artificial Intelligence (中科院二区, IF=8.0, 通讯作者).
- 2024.03: 🎉🎉 New Journal Paper on Novel Pipeline Integrating Cross-Modality and Motion Model for Nearshore Multi-Object Tracking in Optical Video Surveillance Accepted by IEEE Transactions on Intelligent Transportation Systems (中科院一区, Top, IF=8.5, 第一作者).
- 2023.03: 🎉🎉 New Journal Paper on A coarse aggregate gradation detection method based on 3D point cloud Accepted by Construction and Building Materials (中科院一区, Top, IF=7.6, 通讯作者).
- 2023.02: 🎉🎉 New Journal Paper on Foreign Bodies Detector Based on DETR for High-Resolution X-Ray Images of Textiles Accepted by IEEE Transactions on Instrumentation and Measurement (中科院二区, Top, IF=5.6, 第一作者).
- 2023.01: 🎉🎉 New Journal Paper on Sw-YoloX: An anchor-free detector based transformer for sea surface object detection Accepted by Expert Systems With Applications (中科院一区, Top, IF=8.6, 第一作者).
📝 Publications
🎓 Low-level tasks
Modal-Invariant Progressive Representation for Multimodal Image Registration
Jiangang Ding, Yuanlin Zhao, Lili Pei, Yihui Shan, Yiquan Du, Wei Li
| SCI Journal Paper
- Establish a modality-invariant domain to achieve end-to-end registration.
- Propose a progressive dense registration scheme that aligns from coarse to fine.
- Decouple the frequency of multimodal features for independent registration.
- Achieved visually impressive performance, advancing subsequent downstream tasks.
LG-Diff: Learning to Follow Local Class-Regional Guidance for Nearshore Image Cross-modality High-Quality Translation
Jiangang Ding, Yiquan Du, Wei Li, Lili Pei, Ningning Cui
Project | SCI Journal Paper
- Nearshore cross-modal translation is formulated as a denoising diffusion process.
- We develop a two-stage pipeline guided by local class-regional information.
- Decoupling high- and low-frequency information in the feature extraction.
- A cross-modal benchmark for nearshore image translation task is proposed.
🎓 High-level tasks
SeaTrack: Rethinking Observation-Centric SORT for Robust Nearshore Multiple Object Tracking
Jiangang Ding, Wei Li, Ming Yang, Yuanlin Zhao, Lili Pei, Aojia Tian
SCI Journal Paper
- We implemented a fully motion information-based pipeline for the NMOT challenge.
- Solve long-term occlusion problems through guiding modulation between objects.
- Confidence, height and corner information are used to reconstruct the tracker.
- Reduce tracking errors through long and short-term tracklets fusion.
Novel Pipeline Integrating Cross-Modality and Motion Model for Nearshore Multi-Object Tracking in Optical Video Surveillance
Jiangang Ding, Wei Li, Lili Pei, Ming Yang, Aojia Tian, Bo Yuan
Project | SCI Journal Paper
- We propose a holistic cross-modality NMOT pipeline based on optical video surveillance and demonstrate its state-of-the-art performance in multiple benchmark evaluations.
- We developed a simple yet effective CBT to decouple and guide the fusion of the RGB and thermal infrared video streams. Additionally, we introduce a Temporal Context Transformer to mitigate the various complex degradations caused by video signals by integrating feature information from the current and historical frames.
- We heuristically model the motion interrelationships of objects to compensate for the Kalman Filter prior estimates, synchronously combining low-confidence modeling and MCM to assist with association and tracklet correction.
🎖 Honors and Awards
- 2024.10: First-Class Scholarship for Ph.D. Students (Top 10%)
- 2024.10: National Scholarship for Ph.D. Students (Top 1%)
- 2023.10: Excellent Ph.D. Student (Top 10%)
- 2022.10: First-Class Scholarship for Master’s Students (Top 10%)
- 2021.10: Excellent Master’s Student (Top 10%)
- 2021.10: First-Class Scholarship for Master’s Students (Top 10%)
- 2021.05: First Prize in the Provincial “Challenge Cup” Competition (Ranked 2nd)
💻 Projects
- 2024.01 - 2026.01, Fundamental Research Funds for the Central Universities (中央高校优博专项), 主持
- 2023.12 - 2023.12, Scientific Innovation Practice Project of Postgraduates of Chang’an University, 主持
- 2020.09 - 2023.12, National Natural Science Foundation, 参与, 核心人员
- 2023.06 - 2024.12, National Key R&D Program, 参与, 核心人员
💬 Part-time Academic Job
Member of the China Society of Image and Graphics, Member of the Chinese Association for Artificial Intelligence, IEEE Student Member, CSDN Blog Expert (1 million views, 30,000 followers). Long-term reviewer for top journals and conferences in the field of Computer Science.
📖 Educations
- September 2016 - June 2020: Bachelor’s in Measurement and Control Technology and Instruments, Chang’an University, Xi’an, China
- September 2020 - June 2022: Master’s in Information and Communication Engineering, Chang’an University, Xi’an, China, Supervised by Professor Wei Li
- September 2022 - June 2026 (now): Ph.D. in Traffic Information and Engineering, Chang’an University, Xi’an, China, Supervised by Professor Wei Li