YuFu is proactive, love learning new things, and enjoy exploring different ways of thinking. As an R&D software engineer, he has four end-to-end AI development experiences, inclusive of sleep health tracking, body pose tracking, hand gesture recognition, and optics lens design. These experiences consist of co-working with various types of engineers and experts, learning domain knowledge, analyzing data, building up systems, adapting novel techniques to the projects for optimization.積極的で、新しいことを学ぶのが好きで、さまざまな思考方法を探求するのを楽しんでいます。R&Dソフトウェアエンジニアとして、エンドツーエンドのAI開発経験を4つ持っており、睡眠健康診断、ボディポーズの追跡、手のジェスチャー認識、光学レンズ設計を含んでいます。これらの経験には、さまざまなエンジニアや専門家との協力、ドメイン知識の習得、データの分析、システムの構築、新しい技術をプロジェクトに適用して最適化することが含まれます。
-
R And D Software EngineerAlitecsYokohama, Kanagawa, Jp -
Software EngineerAlitecs Dec 2024 - PresentKanagawa, Japan -
R&D Software EngineerAnax Optics Inc May 2024 - Aug 2024Kanagawa, JapanDevelop an optics software which only needs a few clicks to generate commercial lens designs. Moreover, it aims to assist optics engineers at work. Lens Design Extrapolating ● Processed and augmented more than two hundred kinds of lens sequences to a large dataset ● Utilized semi-supervised learning and Deep Learning models to generate lens designs (regression task) ● Customized unsupervised loss function with Ray Tracing ● Adapted automatic hyper-parameter tuning to model training (AutoML) ● Created tools for lens system visualization and evaluation ● The RMS spot size of generated lens design achieved 0.0001 -
R&D Software EngineerHtc Vive Jun 2022 - Apr 2024Taipei, TaiwanWork on ML/DL algorithms research and development in VR/AR and adapt them to our products with Python for real-time applications. Also use C++ for micro-controller/sensor development and computer networking.Body Pose Tracking (Human Motion Estimation)• Processed AMASS datasets and transformed to axis angle, rotation matrix, and 6D representation.• Utilized Text2Motion to synthesize motion clips by text descriptions for data augmentation.• Used a generative model (Diffusion Model) to predict pose (rotation and orientation).• Adapted physics features for optimizing our model performance.• Blended predicted poses with an SMPL model to visualize avatar with motions.Hand Gesture Recognition • Integrated EMG and IMU sensors into the Arduino, and transmitted data via TCP/IP or BLE.• Collected, labeled, pre-processed, and visualized 40+ subjects' data.• Utilized a MobileNet to reach a testing accuracy of 97% for four-gesture classification.• Used transfer learning and few-shot learning to customize new gestures.• Applied for a patent for our hand gesture recognition mechanism and applications.Conducted some tests and experiments for HMD system development by using tool scripts and ADB
Yu-Fu Liu Education Details
-
Electrical And Electronics Engineering -
Electrical And Electronics Engineering
Frequently Asked Questions about Yu-Fu Liu
What company does Yu-Fu Liu work for?
Yu-Fu Liu works for Alitecs
What is Yu-Fu Liu's role at the current company?
Yu-Fu Liu's current role is R and D Software Engineer.
What schools did Yu-Fu Liu attend?
Yu-Fu Liu attended National Cheng Kung University, Chang Gung University.
Free Chrome Extension
Find emails, phones & company data instantly
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial