Salient Object Detection: A Discriminative Regional Feature Integration Approach

Huaizu Jiang¹     Jingdong Wang²    Zejian Yuan¹     Yang Wu³     Nanning Zheng¹     Shipeng Li²

¹Xi'an Jiaotong University     ² Microsoft Research Asia     ³ Kyoto University

Abstract

Salient object detection has been attracting a lot of interest, and recently various heuristic computational models have been designed. In this paper, we regard saliency map computation as a regression problem. Our method, which is based on multi-level image segmentation, uses the supervised learning approach to map the regional feature vector to a saliency score, and finally fuses the saliency scores across multiple levels, yielding the saliency map. The contributions lie in two-fold. One is that we show our approach, which integrates the regional contrast, regional property and regional backgroundness descriptors together to form the master saliency map, is able to produce superior saliency maps to existing algorithms most of which combine saliency maps heuristically computed from different types of features. The other is that we introduce a new regional feature vector, backgroundness, to characterize the background, which can be regarded as a counterpart of the objectness descriptor. The performance evaluation on several popular benchmark data sets validates that our approach outperforms existing state-of-the-arts.

Paper

Huaizu Jiang, Jingdong Wang, Zejian Yuan, Yang Wu, Nanning Zheng, Shipeng Li. Salient Object Detection: A Discriminative Regional Feature Integration Approach. CVPR, 2013. [PDF][Supplementary Material (coming soon)][Bibtex]

Downloads

• Matlab code
• Pixel-wise annotation of MSRA-B Salient Object Dataset
  Split in our CVPR paper: training set, validation set, testing set
• Saliency maps of our approach on 5 benchmark data sets: MSRA-B (on testing set), SED1, SED2, SOD, and iCoSeg

Results

From left to right: input images, results of SVO [1], CA [2], CBsal [3], RC [4], SF [5], LRK [6], and our approach (DRFI), respectively. Please see the paper for details.

Quantitative Comparison

MSRA-B
SED1
SED2
SOD
iCoSeg

Quantitative comparisons of different approaches on benchmark data sets. From left to right: precision-recall curves, ROC curves, and AUC scores, respectively. Please see the paper for details.

References

[1] K.-Y. Chang, T.-L. Liu, H.-T. Chen, and S.-H. Lai. Fusing generic objectness and visual saliency for salient object detection. ICCV, 2011.
[2] S. Goferman, L. Zelnik-Manor, and A. Tal. Context-aware saliency detection. CVPR, 2010.
[3] H. Jiang, J. Wang, Z. Yuan, T. Liu, N. Zheng, and S. Li. Automatic salient object segmentation based on context and shape prior. BMVC, 2011.
[4] M.-M. Cheng, G.-X. Zhang, N. J. Mitra, X. Huang, and S.M. Hu. Global contrast based salient region detection. CVPR, 2011.
[5] F. Perazzi, P. Krahenbuhl, Y. Pritch, and A. Hornung. Saliency filters: Contrast based filtering for salient region detection. CVPR, 2012.
[6] X. Shen and Y. Wu. A unified approach to salient object detection via low rank matrix recovery. CVPR, 2012.

Funding

This work was supported in part by the National Basic Research Program of China under Grant No. 2012CB316400, and the National Natural Science Foundation of China under Grant No. 91120006.

Contact: jianghuaizu AT gmail.com