Lei Chen

I am Lei Chen, an incoming M.S. student in Artificial Intelligence at the Institute of Computing Technology, Chinese Academy of Sciences / University of Chinese Academy of Sciences. Before that, I received my B.Eng. degree in Intelligent Manufacturing Engineering from Southwest Jiaotong University.

My current interests focus on efficient deep learning systems for visual perception, especially stereo matching, model compression, and lightweight deployment. I have worked on optimizing RAFT-Stereo for binocular stereo matching, with an emphasis on practical inference efficiency, FP16 deployment, and edge-oriented acceleration.

Beyond computer vision, I am also interested in large language models, embedding models, retrieval-augmented generation, and AI-assisted developer tools. I enjoy building self-hosted systems, automation tools, and practical AI applications that connect research ideas with real-world workflows.

中文简介：我目前是中国科学院大学 / 中国科学院计算技术研究所人工智能方向专业硕士新生，本科毕业于西南交通大学智能制造工程专业。我的研究兴趣包括双目立体匹配、高效深度学习、多模态感知、大语言模型与检索增强生成等方向。我希望将深度学习模型从“能跑”进一步推进到“高效、稳定、可部署”。

Research Interests

• Efficient Deep Learning and Model Compression
• Stereo Matching and 3D Vision
• Multimodal Perception
• Large Language Models and RAG
• Representation Learning and Embedding Models
• AI-powered Automation Tools

Selected Projects

• RAFT-Stereo Optimization and Lightweight Deployment — optimizing binocular stereo matching for practical FP16 inference and deployment-oriented acceleration.
• Competitive Price Monitoring System for E-commerce — Playwright-based automation for SKU-level price extraction and discount-aware comparison.
• Self-hosted Personal Infrastructure — personal web systems and self-hosted services around the somnus.top domain.

Manuscripts / Preprints

No formal publications yet. I currently keep research notes and thesis-related work in progress, including an undergraduate thesis on lightweight binocular stereo matching.