Research Projects

Annotating Gigapixel Images

    Panning and zooming interfaces for exploring very large images containing billions of pixels (gigapixel images) have recently appeared on the internet. This paper addresses issues that arise when creating and rendering auditory and textual annotations for such images. In particular, we define a distance metric between each annotation and any view resulting from panning and zooming on the image. The distance then informs the rendering of audio annotations and text labels. We demonstrate the annotation system on a number of panoramic images.  (Paper, Video Page)

Natural Image Colorization


    In this paper, we introduce an interactive system for colorizing natural images of complex scenes. Previous interactive colorization methods based on intensity-continuity cannot easily colorize highly textured regions in natural images. We present a novel color labeling scheme that integrates intensity-continuity and texturesimilarity
constraints in a natural image: not only spatially neighboring pixels with similar intensities, but also spatially distant pixels with similar texture should have similar colors. With a few color scribbles, ... (Project page)


 Color Transfer Brush

    In this paper, we introduce an interactive tool for local color transfer. The new technique is based on the observation that color transfer operations are local in nature while at the same time should adhere global consistency. We introduce a brush by which the user specifies the source and destination image regions for color transfer. Color statistics in the source region are transferred to the destination region. A global optimization is then applied to eliminate visual discontinuities that may result by the local operations. We demonstrate that our tool is easy to use yet effective in quickly generating diverse artistic effects. (Project page)


 Color Sketch Generation


In this paper, we introduce an NPR system to generate color sketches in a free-hand drawing style. From observation, we found out that in a free-hand style both shape and color of image regions are stylized, which contributes cooperatively to the visual effect of a sketch. With the color regions obtained by our interactive segmentation algorithm, we create a stylized color sketch in two steps. First, in order to manifest the lighting and shading of each region, we shrink the boundary of the color regions with a luminance-based algorithm emphasizing the highlight parts,... (Project page)

Automatic Pixeldraw Generation

 Pixel draw is a traditional drawing approach for screen, which is reviving recently with the development of Internet and cell phone. Now it has become an important art form in the realm of illustration. However, drawing pixel draw is a complex work, which is effort and time consuming. Thus, generating pixel draw from photo becomes an attractive researching topic. In this article, we propose a novel approach to automatically generate facial pixel draw based on the face features detect by ASM. The algorithm divides the face into several facial components; each component is then processed individually. A new concept, “Color Palette”, is introduced in our system, which is the key in creating satisfactory pixel draw effect. We demonstrate the effectiveness of our method in a number of photo generated results. (Project page: Coming Soon)