Experience

NeRF based novel-view synthesis with a focus on photorealistic rendering and reposing of moving subjects with sparse views.

Learning to anticipate human actions that happen in the next one to five seconds using self-regularized future content generation.

Object detection, recognition and scene understanding for self-driving vehicles and underground imaging.

Tree-structured convolutional and capsule networks for code analysis and related NLP tasks.

Computer vision and image processing techniques for video surveillance.

Research objectives included designing and deploying a massive scale non-invasive behavior monitoring system inside public buildings, capturing the mobility and behavioral pattern of the occupants, and predicting the occupants' mental well-being using deep learning and machine learning techniques.