Responsibilities include:

• Designing and implementing VR applications in Unity3D, leveraging spatial computing and advanced motion planning on Oculus Quest to decode user behavior and cognitive states through motion dynamics.
• Utilizing cutting-edge machine learning techniques to build intelligent systems that interpret and anticipate subtle cognitive states, advancing seamless human-computer interaction.
• Conducting independent research initiatives encompassing participant recruitment, structured data collection, deep learning experimentation, and comprehensive result analysis.

TCTM offers quality IT education targeting students aged three to eighteen through a hybrid platform combining live instruction, classroom tutoring, and online learning. The company was founded in 2002, formerly known as Tarena International, Inc.

Responsibilities include:

• Led instruction for two intensive training camps, delivering seven classes to 210+ students aged 9 to 18, featuring an engaging, clear, and accessible teaching style.
• Designed and taught a comprehensive curriculum spanning foundational C++ syntax to advanced algorithms and data structures, integrating cutting-edge computer technologies to inspire young students and cultivate their problem-solving skills.

Responsibilities include:

• Independently led a computer vision project that leveraged laser-illuminated wrist videos to estimate heart rate, integrating traditional digital image processing techniques with deep learning models to improve the accuracy of physiological measurements.
• Investigated accelerating Neural Radiance Field (NeRF) training using PyTorch, improving computational efficiency in 3D scene reconstruction.

Responsibilities include:

• Researched on cervical spondylosis measurement using a U-Net-based model in PyTorch for segmentation and MMPose for pose estimation.
• Achieved a 75% reduction in Cobb metric error (from 20° to 4.88°) through advanced image processing and model optimization, demonstrating expertise in computer vision and machine learning.
• Collaborated with clinicians from Peking University Third Hospital to ensure accurate data labeling and validation.
• Advisor: Prof. Xinhang Song