Research Areas
Image and video segmentation and editing, object classification, image-based modeling, computer vision, and image processing.

Research Description
With the increasing use of digital images and video in web-based content, entertainment, medical analysis, virtual environment creation, etc., the demand will continue to grow for fast and easy-to-use tools that accurately select, extract, measure, filter, and/or otherwise manipulate objects of interest within images and image sequences. One key step in many image editing/manipulation/quantification applications is segmentation (or specifying the region or regions of interest). For this reason, research continues to look for better selection tools that can reduce the human burden involved in general-purpose segmentation applications.

My research focuses on developing tools and techniques for accurate and effective user-guided segmentation and the subsequent measurement, extraction, filtering, composition, and/or other quantification/manipulation of objects within digital images and image sequences (such as videos and medical image volumes). Included in this research is the continued improvement of intelligent scissors, a general purpose selection tool, as well as improved methods for automatic matte extraction, automatic background and foreground color separation, subpixel estimation of object boundaries, and automatic contour tracking in temporal or spatial sequences.

Applications of Research
This work has potential benefit several areas including medical image analysis (both 2-D and 3-D), image-based modeling and rendering, virtual environment creation, and many other applications where efficient and accurate object selection is required. My intelligent scissors research has already been incorporated into some publicly available image editing and manipulation packages such as Adobe Photoshop and Photoshop Elements (as the magnetic lasso tool) and GIMP (Gnu Image Manipulation Package).