Curriculum Vitae
Education
Southern University of Science and Technology
- GPA: 3.71/4.0. Major rank: 78/152, calculated from the cumulative GPA of students in the same cohort and major.
- Selected coursework: Computer Vision 97 (A+), Artificial Intelligence 94 (A), Design and Analysis of Algorithms 94 (A), Information Retrieval 97 (A+).
- English: CET-4 615, CET-6 515, IELTS 7.0 (Listening 6.5, Reading 8.5, Writing 6.0, Speaking 6.0).
Research interests
Robotics vision, real-time visual perception, edge visual-model deployment, vision-language models, multimodal retrieval, and efficient visual models. I hope to study efficient, robust, and deployable visual intelligence for real-world settings.
Research training and project experience
See the Research & Projects page for technical detail.
Experimental Study on Multi-turn Multimodal Fashion Image Retrieval
Contributing to FashionAM experimental validation, training evaluation, and ablation analysis across text-to-image, composed image retrieval, in-shop retrieval, and sketch-to-image tasks in multi-turn settings.
Experimental Study on Diffusion Model Acceleration and Artifact Analysis
Implemented and analyzed FLUX.1-dev experiments comparing the effects of VAE re-projection on high-frequency artifacts, image quality, and inference efficiency.
RoboMaster Robotics Competition / ARTINX Lab
Worked on model training, inference deployment, and on-robot debugging for hero, infantry, and sentry robot vision systems.
VLM Image Captioning and Component Analysis
Built and evaluated a LLaVA-style vision-language model with PyTorch and Transformers, including component ablations.
Face Recognition and Expression Interaction System on Jetson Nano
Built a real-time face-detection, face-recognition, and expression-interaction system on a Jetson Nano 4GB board, achieving about 30 fps end-to-end.
Technical skills
- Programming and engineering: C++, Python, Git, Linux/Windows project development.
- Deep learning and multimodal learning: PyTorch, Transformers, MLLM, LoRA, contrastive learning, retrieval evaluation.
- Vision and deployment: YOLO, RetinaFace, MobileFaceNet, MobileNetV3, TensorRT, Jetson Nano, Jetson Orin NX, ROS.
- Systems and algorithms: C++ Actor Framework, Extended Kalman Filter, real-time vision-system debugging.
Honors and awards
- National Encouragement Scholarship, Dec. 2025.
- Outstanding Student, Southern University of Science and Technology, Dec. 2025.
- Team Second Prize, RoboMaster 2025 University Championship, Aug. 2025.
- Team First Prize, RoboMaster 2025 Southern Regional Competition, Jul. 2025.
- Outstanding Student, Southern University of Science and Technology, Dec. 2024.
- Third-Class Outstanding Student Scholarship, Southern University of Science and Technology, Dec. 2024.