Learning Optimal Seeds for Ranking Saliency
A variety of methods have been developed for visual saliency analysis, and it is a challenge to detect the most important scene from the input image. In this paper, to improve the shortage that the spatial connectivity of every node in model only via the k-regular graph and the idealistic boundary prior assumption is used in graph-based manifold ranking, we present a new optimal seed method to get saliency map. First, we evaluate the salience value of each region by global contrast-based spatial and color feature. Second, the salience values of the first stage are used to optimize the background and foreground queries (seeds); meanwhile, we tackle boundary cues from hierarchical graph to optimize background seeds. Then, we derive each stage saliency measure by the classical manifold ranking after obtaining optimal seeds. Finally, the final saliency map is obtained by combining the saliency results of two stages. Our algorithm is tested on the five public datasets and compared with nine state-of-the-art methods; the quantitative evaluation indicates that our method is effective and efficient. Our method can handle complex images with different details and can produce more accurate saliency maps than other state-of-the-art approaches.
KeywordsSaliency detection Global region contrast Hierarchical graph Optimal seeds Manifold ranking
This work was supported by the grant of the National Natural Science Foundation of China (61671018, 61672204), Major Science and Technology Project of Anhui Province (17030901026), Key Constructive Discipline Project of Hefei University (2016xk05), MOE Youth Project of Humanities and Social Sciences (15YJC860034), Natural Science Foundation of Anhui Higher Education Institutions (KJ2016A604), Key Project of Hefei University (16ZR21ZDA), Youth Backbone Visiting Research Key Project at Abroad (gxfxZD2016219), National Statistical Science Research Project of China (2014LZ32), Horizontal Cooperative Research Project of Fuyang Normal University (XDHX2016021).
This study was funded by the National Natural Science Foundation of China (Grant No. F011601), National Nature Science Foundation of China (No. 61672204), Humanity and Social Science Youth Foundation of Ministry of Education of China (Grant No. 15YJC860034), Natural Science Foundation of Anhui Province of China (Grant No. KJ2016A604), and Natural Science Foundation of Fuyang Normal University (Grant No. 2017FSKJ11).
Compliance with Ethical Standards
Conflict of Interest
The authors declare that they have no conflict of interest.
This article does not contain any studies with human participants or animals performed by any of the authors.
- 1.Xie YL, Lu HC, Yang MH. Bayesian saliency via low and mid-level cues. IEEE Trans Image Process. 2013;5(22):1689–98.Google Scholar
- 2.Shen X and Wu Y. A unified approach to salient object detection via low rank matrix recovery. In CVPR; 2012.Google Scholar
- 3.Achanta R, SHemami S, Estrada FJ, and Süsstrunk S. Frequency-tuned salient region detection. CVPR; 2009. p. 1597–1604.Google Scholar
- 4.Y. Zhai and M. Shan. Visual attention detection in video sequences using spatiotemporal cues. In ACM Multimedia; 2006. p. 815–824.Google Scholar
- 6.Palmer S. Vision science: photons to phenomenology, volume 1. Cambridge: MIT; 1999.Google Scholar
- 7.Klein DA and Frintrop S. Center-surround divergence of feature statistics for salient object detection. ICCV; 2011.Google Scholar
- 10.Borji A, Cheng MM, Jiang H, Li J. Saliency object detection: a survey. arXiv preprint arZXiv. 2014;1411–5878.Google Scholar
- 12.Lee YJ, Ghosh J, and Grauman K. Discovering important people and objects for egocentric video summarization. CVPR; 2013. p. 1346–1353.Google Scholar
- 13.Perazzi F, Krahenbuhl P, Pritch Y, and Hornung A. Saliency filters: contrast based filtering for salient region detection. In Proc. IEEE Conf. Comput. Vis; 2011. p. 733–740.Google Scholar
- 14.M. Cheng, J. Warrell, W. Lin, S. Zheng, V. Vineet, and N. Crook, Efficient salient region detection with soft image abstraction. In Proc. IEEE Conf. Comput, Vis; 2013. p. 1529–1536.Google Scholar
- 15.Yang C, Zhang L, Lu H, Ruan X, and Yang MH. Saliency detection via graph-based manifold ranking. In CVPR; 2013. p. 3166–3173.Google Scholar
- 16.W. Wang, Y Wang, Q. Huang, and W. Gao. Measuring visual saliency by site entropy rate. In Proc IEEE Conf Comput Vis Pattern Recog; 2010. p. 2368–2375.Google Scholar
- 18.Bruce N and Tsotos JK. Saliency based on information maximization. NIPS; 2006. p. 155–162.Google Scholar
- 21.Ma YF and Zhang HJ. Contrast-based image attention analysis by using fuzzy growing. In ACM International Conference on Multimedia; 2003. p. 374–381.Google Scholar
- 22.Yan Q, Xu L, Shi JP, and Jia JY. Hierarchical saliency detection. In CVPR; 2013. p. 1155–1162.Google Scholar
- 23.Harel J, Koch C, Perona P. Graph-based visual saliency. In Advances in Neural Information Processing Systems. 2006. p. 545–552.Google Scholar
- 25.Huang Y, Wang W, Wang L and Tan T. An effective regional saliency model based extended site entropy rate. In ICPR; 2012. p. 1407–1410.Google Scholar
- 26.Jiang Z and Davis L. Submodular salient region detection. In Proc IEEE Conf Comput Vis Pattern Recog; 2013. p. 2043–2050.Google Scholar
- 27.Jiang P, Ling H, Yu J, and Peng J. Salient region detection by UFO: uniqueness, focusness and objectness. In ICCV; 2013.Google Scholar
- 28.Jiang BW, Zhang LH, Lu HC. Saliency detection via absorbing markov chain. Proc IEEE Int Conf Comput Vis Sydney. 2013;2013:1665–72.Google Scholar
- 32.Perazzi F, Krahenbuhl P, Pritch Y, and Hornung A. Saliency filters: contrast based filtering for salient region detection. In CVPR; 2012. p. 733–740.Google Scholar
- 33.Hou X and Zhang L. Saliency detection: a spectral residual approach. In IEEE CVPR; 2007. p. 1–8.Google Scholar
- 34.Murray N, Vanrell M, Otazu X, and Parraga CA. Saliency estimation using a non-parametric low-level vision model. In IEEE CVPR; 2011. p. 433–440.Google Scholar