Bio

My current research focuses on using deep learning techniques on image data, which can be obtained from the Internet, UAVs, or satellites, to improve object recognition and scene understanding in the wild with a particular focus on applications in the social good domain ranging from disaster response and management to sustainable development.

In the past, I worked on several topics such as creating a holistic view of an individual's health, food image and recipe understanding, human body motion analysis and action recognition from 3D data, multimodal dance performance analysis and music-driven dance synthesis and animation as well as facial expression analysis and speech-driven facial expression synthesis and animation.