Field of Expertise
deep learning, intelligent robotics, and image/video/audio processing
Expertise
I work in the domain of deep learning, intelligent robotics, and image/video/audio processing, for extreme and challenging environments, engineering defect, health and biodiversity monitoring, medical imaging, underwater vehicle and other video/audio-based applications. More detailed applications include video action recognition, deepfake detection, person re-identification, biometrics detection (via retina/finger print/palm print/face images), mental health detection, video/audio summary, video description generation, acoustic event prediction, audio-based diabetes, heart disease, dementia, and lung condition detection, extreme weather event prediction and rescue. Novel deep networks and weakly supervised learning methods have been developed to tackle the above real-world challenges, which have great potential in leading to various social and societal impact.