My research interests are in the general areas of pattern recognition, computer vision, machine learning with human-computer interactions, and remote sensing. The current projects that I am working on include hyperspectral image processing, contents-based image retrieval, graph model for object matching/classification/retrieval, and interactive image annotation.
Recently, I have won the Discovery Early Career Researcher Award (DECRA) from the Australian Research Council (ARC) for a projec titled "Interactive Computer Vision for Image Interpretation". Image interpretation refers to the reconstruction and interpretation of a scene by means of images. Interactive computer vision studies the human perception process and the interactions between human and the image interpretation systems. Besides the tasks traditionally performed by computer vision, such as feature extraction, object detection, and scene understanding, the proposed research focuses more on developing a framework in which the computer vision system learns from human in an image interpretation environment, and gradually optimises its performance until the desired outcomes are achieved. This project aims at pushing forward the fundamental research in interactive computer vision, and developing practical solutions for real-world image interpretation tasks.
The topics to be investigated in this project include: