Summary
Lex Song is an independent machine learning researcher with over a decade of hands-on experience in computer vision, video tracking, and face image processing, spanning synthesis, analysis, expression transfer, and identity swapping. He combines deep-learning prototyping with production-ready C and modern C++ engineering, plus Python scientific stack expertise and cross-language integration. Previously a senior ML scientist at Digital Domain, he led projects in TTS-driven facial animation and 3D face reconstruction, and earlier improved VR tracking algorithms at HTC Vive. Comfortable across problem domains and platforms, he’s also reverse-engineered geospatial tiling logic for an open-source project mapping Google’s 3D satellite mode—an example of his curiosity for low-level systems as well as high-level models. Based in Taipei, he’s known for rapid adaptation and turning novel algorithms into practical demos and tools.
10 years of coding experience
11 years of employment as a software developer
Master's degree, Computer Science, Master's degree, Computer Science at National Taiwan University
English, Chinese, Japanese