M.Sc. student in AI (GPA 18.45/20), preparing for thesis defense. Research on uncertainty-guided learning in visual speech recognition and decision making in Go. Seeking a fully funded PhD in computer vision or efficient deep learning.
EGOSD uses an auxiliary belief module to estimate frame-level uncertainty and control online self-distillation in a lightweight VSR model, with negligible additional inference cost.
An open-source Python package that extracts lip regions from video with MediaPipe. It supports temporal smoothing, optional illumination normalization, and configurable processing on CPU or CUDA.