- From Annotation Tools to Real-World AI: My Open-Source Journey with Labelme, Medical Imaging, and Industrial Vision / 綜合議程 - 各種開源議題 Page
- 描述 In practical artificial intelligence development, while model design is important, project success often depends more on data annotation workflows, toolchain integration, and engineering deployment. This talk draws on my real-world contributions to the open-source annotation tool Labelme, showing how practical user needs led to multilingual localization and feature improvements, as well as the technical trade-offs involved in collaborating with upstream maintainers and the community. Building on this experience, the session explores how open-source tools help bridge AI research and real-world applications, highlighting challenges such as data quality and privacy in medical image segmentation and data distribution shifts and real-time requirements in multi-site industrial vision deployment. Through cross-domain case studies, it emphasizes the critical role of open-source tooling in modern AI workflows and demonstrates how addressing real problems through community collaboration can effectively support engineering practice in medical and industrial AI systems.
- 整合開源與學術成果：構建隱私保護的人工智能框架方法 / COSCon'25 第十届中国开源年会北京
- 描述 Integrating open-source technologies with academic research findings to build an artificial intelligence framework that balances privacy protection and model performance. / 整合開源技術與學術研究成果，構建在隱私保護與模型效能之間取得平衡的人工智能框架。
- Beginner AI Experiments: Practicing AI with Open-Source Resources Video, Page
- 描述 As a graduate student with a background in computer science, my journey did not begin with artificial intelligence or medical image segmentation. Initially, I often felt overwhelmed and anxious about the steep technical barriers posed by deep learning and model training. However, thanks to the open-source community and the wealth of available tools, I had the opportunity to start from scratch and gradually build my own AI research project. In this talk, I will share my journey from the perspective of a newcomer without a formal background in AI. I will explain how I leveraged public medical datasets, the PyTorch framework, and open-source models on GitHub—such as Vision Mamba UNet—to develop a practical, research-ready medical image segmentation system. By reading documentation, studying open-source code, and participating in developer communities, I gained valuable hands-on experience and learned how to complete a research project even with limited resources. This session focuses on real-world insights such as “how to get started with AI through open-source,” “the challenges and lessons learned from building my first project,” and “common mistakes beginners should avoid.” I hope my experience will encourage more students and developers who are curious about AI but have not yet taken the first step: you do not need to build everything from scratch. By making the most of open-source resources, you can go farther—and do so with greater confidence—on your AI journey.
