Automatic foreground segmentation using light field images

Foreground segmentation is a fundamental method in computer vision. Traditional foreground segmentation algorithms are sensitive to blurry degree of background, smooth foreground regions and camouflage foreground. To deal with these problems, we use light field images as input by exploiting its focusness cue. In this paper, we propose an automatic foreground segmentation algorithm for light field images. Firstly we calculate focusness measures in refocused stack and oversegment all focus images. Then the graph cut framework is introduced to generate foreground result. Experiments show that our method has a superior performance against counterparts and is more robust in scenes like ambiguous foreground.

[1]  Qi Zhao,et al.  A Fuzzy Segmentation of Salient Region of Interest in Low Depth of Field Image , 2007, MMM.

[2]  Vladimir Kolmogorov,et al.  An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Vision , 2004, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Harry Shum,et al.  Lazy snapping , 2004, ACM Trans. Graph..

[4]  Erin Mayhood,et al.  Advances in Multimedia , 2007 .

[5]  Mark Q. Shaw,et al.  Automatic Image Segmentation by Dynamic Region Growth and Multiresolution Merging , 2009, IEEE Transactions on Image Processing.

[6]  Hans-Peter Kriegel,et al.  Robust Image Segmentation in Low Depth Of Field Images , 2013, ArXiv.

[7]  P. Hanrahan,et al.  Light Field Photography with a Hand-held Plenoptic Camera , 2005 .

[8]  Changick Kim,et al.  Segmenting a low-depth-of-field image using morphological filters and region merging , 2005, IEEE Transactions on Image Processing.

[9]  Hans-Peter Kriegel,et al.  Robust segmentation of relevant regions in low depth of field images , 2011, 2011 18th IEEE International Conference on Image Processing.

[10]  King Ngi Ngan,et al.  Unsupervized Video Segmentation With Low Depth of Field , 2007, IEEE Transactions on Circuits and Systems for Video Technology.

[11]  Huijun Gao,et al.  A Curve Evolution Approach for Unsupervised Segmentation of Images With Low Depth of Field , 2013, IEEE Transactions on Image Processing.

[12]  Haibin Ling,et al.  Saliency Detection on Light Field , 2014, CVPR.

[13]  King Ngi Ngan,et al.  Learning to Extract Focused Objects From Low DOF Images , 2011, IEEE Transactions on Circuits and Systems for Video Technology.

[14]  Du-Ming Tsai,et al.  Segmenting focused objects in complex visual images , 1998, Pattern Recognit. Lett..

[15]  Vladimir Kolmogorov,et al.  "GrabCut": interactive foreground extraction using iterated graph cuts , 2004, ACM Trans. Graph..

[16]  Jian Sun,et al.  Lazy snapping , 2004, SIGGRAPH 2004.

[17]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.