Traditional image matting approaches requires user interaction. This paper proposes an automatic framework for natural image matting. The method seamlessly incorporates image matting with the top-down process of segmentation by weighted aggregation to get a rich and multi-scale grapy pyramid representation of the input image. Using the coupling between aggregates in the graph pyramid, the region for matting is detected adaptively and automatically. Meanwhile, foreground and background regions are determined with state variables. An energy function is constructed to represent the similarity and smoothness properties of a matte and is iteratively optimized. Under the automatic matting framework, color sampling is more accurate than existing methods since multi-scale measurements such as intensity and texture are fully considered. Experiments show that the proposed automatic method is more efficient to extract high quality matte even for difficult images in which foreground and background have very similar colors. Another attractive feature of the method is that it can extract mattes for multi-objects at one computing time.
Image matting is a process of extracting a foreground object from a complex background. This paper proposes a robust
interactive image matting approach. The method requires only a few user interactions in the form of drawing a rectangle
and a few strokes to indicate background and foreground. We consider the constraints of accuracy and continuity for the
estimated alpha values together to find the optimal matte by iteratively energy optimization. Different from existing
sampling-based natural image matting methods which use only intensity information from statistic sampling of known
foreground and background pixels to estimate the unknown pixels. We consider the distribution of the known pixels in
color, texture and spatial spaces, and build a more robust statistical model. At each iteration, the statistical model is
updated according to previous results of matting. Furthermore an accuracy function of sampling is proposed. These
manipulations make the sampling of foreground and background pixels more accurate and thus improve the performance
of the matting processing. Experiments show that compared with previous approaches, our method is more efficient to
extract high quality matte for texture-rich images and difficult images in which foreground and background have very
similar colors, while requiring a surprising small amount of user interaction.