This book focuses on interactive segmentation techniques, which have been extensively studied in recent decades. Interactive segmentation emphasizes clear extraction of objects of interest, whose locations are roughly indicated by human interactions based on high level perception. This book will first introduce classic graph-cut segmentation algorithms and then discuss state-of-the-art techniques, including graph matching methods, region merging and label propagation, clustering methods, and segmentation methods based on edge detection. A comparative analysis of these methods will be provided with quantitative and qualitative performance evaluation, which will be illustrated using natural and synthetic images. Also, extensive statistical performance comparisons will be made. Pros and cons of these interactive segmentation methods will be pointed out, and their applications will be discussed.
There have been only a few surveys on interactive segmentation techniques, and those surveys do not cover recent state-of-the art techniques. By providing comprehensive up-to-date survey on the fast developing topic and the performance evaluation, this book can help readers learn interactive segmentation techniques quickly and thoroughly.
|
|
|
1 | (6) |
|
|
|
3 | (4) |
|
2 Interactive Segmentation: Overview and Classification |
|
|
7 | (10) |
|
|
|
7 | (2) |
|
2.2 Graph Modeling and Optimal Label Estimation |
|
|
9 | (3) |
|
2.3 Classification of Solution Techniques |
|
|
12 | (5) |
|
|
|
15 | (2) |
|
3 Interactive Image Segmentation Techniques |
|
|
17 | (46) |
|
|
|
17 | (15) |
|
|
|
18 | (1) |
|
3.1.2 Interactive Graph-Cut |
|
|
19 | (2) |
|
|
|
21 | (2) |
|
|
|
23 | (2) |
|
|
|
25 | (3) |
|
3.1.6 Graph-Cut with Prior Constraints |
|
|
28 | (3) |
|
3.1.7 Multi-Resolution Graph-Cut |
|
|
31 | (1) |
|
|
|
31 | (1) |
|
3.2 Edge-Based Segmentation Methods |
|
|
32 | (6) |
|
|
|
32 | (1) |
|
3.2.2 Live-Wire Method and Intelligent Scissors |
|
|
33 | (3) |
|
3.2.3 Active Contour Method |
|
|
36 | (1) |
|
|
|
37 | (1) |
|
|
|
38 | (8) |
|
|
|
38 | (5) |
|
3.3.2 Random Walk with Restart (RWR) |
|
|
43 | (3) |
|
|
|
46 | (1) |
|
|
|
46 | (10) |
|
3.4.1 Pre-Processing for Region-Based Segmentation |
|
|
46 | (2) |
|
3.4.2 Seeded Region Growing (SRG) |
|
|
48 | (1) |
|
|
|
49 | (1) |
|
3.4.4 Maximal Similarity-Based Region Merging |
|
|
50 | (2) |
|
3.4.5 Region-Based Graph Matching |
|
|
52 | (3) |
|
|
|
55 | (1) |
|
3.5 Local Boundary Refinement |
|
|
56 | (7) |
|
|
|
57 | (6) |
|
|
|
63 | (12) |
|
|
|
63 | (2) |
|
4.2 Evaluation on Challenging Images |
|
|
65 | (6) |
|
4.2.1 Images with Similar Foreground and Background Colors |
|
|
65 | (2) |
|
4.2.2 Images with Complex Contents |
|
|
67 | (1) |
|
4.2.3 Images with Multiple Objects |
|
|
67 | (2) |
|
|
|
69 | (2) |
|
|
|
71 | (4) |
|
|
|
73 | (2) |
|
5 Conclusion and Future Work |
|
|
75 | |
Jia He received her B.S. and M.S. degree in Information and Communication Engineering from Xi'an Jiaotong University (XJTU), China, in 2007 and 2010 respectively. She worked in Realsil Microelectronics (Suzhou) Co. LTD. as system engineer in 2010. In 2011 she joined the Media Communication Lab lead by Professor Kuo in University of Southern California (USC), where she is pursuing her Ph.D degree in Electrical Engineering and serving as research assistant. Her research interests include image/video processing and computer vision.
Chang-Su Kim received the Ph.D degree in electrical engineering from Seoul National University (SNU) with a Distinguished Dissertation Award in 2000. From 2003 and 2005, he was an Assistant Professor in the Department of Information Engineering, Chinese University of Hong Kong. In Sept. 2005, he joined the School of Electrical Engineering, Korea University, where he is now a Professor. His research topics include image processing and multimedia communications. In 2009, he received the IEEK/IEEE Joint Award for Young IT Engineer of the Year. He has published more than 170 technical papers in international journals and conferences. He is a Senior Member of IEEE. He is an Editorial Board Member of Journal of Visual Communication and Image Representation and an Associate Editor of IEEE Transactions on Image Processing.
C.-C. Jay Kuo Dr. C.-C. Jay Kuo received the B.S. degree from the National Taiwan University, Taipei, in 1980 and the M.S. and Ph.D. degrees from the Massachusetts Institute of Technology, Cambridge, in 1985 and 1987, respectively, all in Electrical Engineering. From October 1987 to December 1988, he was Computational and Applied Mathematics Research Assistant Professor in the Department of Mathematics at the University of California, Los Angeles. Since January 1989, he has been with the University of Southern California (USC). He is presently Director of the Multimedia Communication Lab. and Professor of Electrical Engineering and Computer Science at the USC. His research interests are in the areas of multimedia data compression, communication and networking, multimedia content analysis and modeling, and information forensics and security. Dr. Kuo has guided 115 students to their Ph.D. degrees and supervised 22 postdoctoral research fellows. Currently, his research group at the USC has around 30 Ph.D. students (please visit website http://viola.usc.edu), which is one of the largest academic research groups in multimedia technologies. He is co-author of about 200 journal papers, 850 conference papers and 10 books. He delivered over 550 invited lectures in conferences, research institutes, universities and companies.
