Precise Object Cutout from Images
Ming Liu, Shifeng Chen, and Jianzhuang Liu
In this paper we propose a novel approach to the problem of interactive foreground/background segmentation in images. With user provided strokes which indicate foreground and background seeds, we estimate two Gaussian mixture models, one for foreground and the other for background, and define two quantities to measure the initial probabilities of each pixel belonging to the foreground and the background respectively. An optimization function constructed based on the quantities and the boundary and coherent region information is proposed to solve the segmentation problem. By relaxing the hard binary segmentation to a soft labelling problem in the continuous domain, a closed form global optimal solution can be achieved, which directly results in the final binary segmentation output. Experimental results demonstrate the excellent performance of our algorithm.
Figure 1: Interactive segmentation result by our algorithm. (a) Original image with user-specified strokes (white for foreground and black for background). (b) The map of the probability of pixel belonging to the object region, where the brighter the pixel is, the closer to 1 the probability is. (c) Global optimal result in the continuous domain. (d) Final object extracted.
Figure 2: Results on “Flower” and “Person” images. From left to right: input images with user guided strokes, the results of BP, GC, and our algorithm. We also zoom in some regions for better observation.
Figure 3: Some experimental results obtained by our algorithm.
Table 1: Comparison of the error rates on the 50 images in the database (available at http://research.microsoft.com/vision/cambridge).
ü M. Liu, S. Chen, and J. Liu, “Precise Object Cutout from Images,” Proc. ACM Int. Conf. Multimedia (ACM MM), 2008. [pdf]