🎓 Artificial Intelligence undergraduate at Huazhong University of Science and Technology.
I am interested in computer vision, speech enhancement, multimodal AI, and on-device AI deployment. My current work focuses on applying deep learning methods to practical problems in image restoration, audio restoration, and intelligent recognition.
- Computer Vision and Image Restoration
- Speech Enhancement and Audio Signal Processing
- Multimodal AI Applications
- On-device AI and Model Deployment
- AI-assisted Software Development
Exploring how high-quality reference images can improve facial detail restoration under different poses, expressions, and lighting conditions.
Developing deep-learning-based methods for restoring low-bandwidth and sensor-style speech signals for robust mobile voice communication.
Building a fruit detection and maturity classification pipeline using YOLO and deep convolutional neural networks.
Evaluating and improving image matting models for complex plant images, with a focus on foreground extraction and failure-case analysis.
- Programming: Python, C++, Java, Kotlin
- Deep Learning: PyTorch, CNNs, Transformers
- Computer Vision: OpenCV, YOLO, ResNet, Image Restoration
- Speech Processing: STFT, Speech Enhancement, PESQ, STOI
- Model Deployment: ONNX Runtime, Android
- Development Tools: Git, GitHub, Docker, Codex, Claude Code
- Reference-guided image restoration
- Multimodal feature fusion
- Real-time and on-device AI inference
- AI application development and engineering
- GitHub: MysticIris