Statistical Textural Distinctiveness for Salient Region Detection in Natural Images

A novel statistical textural distinctiveness approach for robustly detecting salient regions in natural images is proposed. Rotational-invariant neighborhood-based textural representations are extracted and used to learn a set of representative texture atoms for defining a sparse texture model for the image. Based on the learnt sparse texture model, a weighted graphical model is constructed to characterize the statistical textural distinctiveness between all representative texture atom pairs. Finally, the saliency of each pixel in the image is computed based on the probability of occurrence of the representative texture atoms, their respective statistical textural distinctiveness based on the constructed graphical model, and general visual attentive constraints. Experimental results using a public natural image dataset and a variety of performance evaluation metrics show that the proposed approach provides interesting and promising results when compared to existing saliency detection methods.

Architecture for salient region detection based on sparse texture modeling and statistical textural distinctiveness

Illustration of the architecture for salient region detection based on sparse texture modeling and statistical textural distinctiveness: Based on textural representations, we learn a sparse texture model and build a graphical model to computed the final saliency map.

Salient objects with texture patterns that are visually significantly different from those  of the rest of the sceneFrom left to right: Salient objects with texture patterns that are visually significantly different from those  of the rest of the scene. Pixels associated with the corresponding atom of a learned texture model.  Computed saliency map and ground truth mask.

Code

 If you are interested in our software for saliency computation based on statistical textural distinctiveness, we would be happy to share demo code (MATLAB) for research purposes only. You may request your demo code by sending an email to: cscharfenberger@uwaterloo.ca

Publication

Christian Scharfenberger, Alexander Wong, Khalil Fergani, John Zelek, and David Clausi, Statistical Textural Distinctiveness for Salient Region Detection in Natural Images, In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2013 [Paper] [Poster] [Bibtex]

Downloads  

.txtcitation.txt

.pdfStatTexSal_CVPR_Scharfenberger.pdf

 Related people

Directors

Alexander Wong, David A. Clausi

Students

Bi Hongbo

Alumni

Christian Scharfenberger

Related research areas

Image Segmentation/Classification

Related publications 

 Journal articles

Scharfenberger, C.A. ChungA. Wong, and D. A. Clausi, "Salient Region Detection Using Self-Guided Statistical Non-Redundancy in Natural Images", IEEE Access, vol. 4, pp. 48-60, 2016. Details

Scharfenberger, C.A. Wong, and D. A. Clausi, "Structure-guided Statistical Textural Distinctiveness for Salient Region Detection in Natural Images", IEEE Transactions on Image Processing, vol. 24, no. 1, pp. 457-470, January, 2015. Details

Conference papers

Karimi, A-H., J. M. Shafiee, C. Scharfenberger, I. B. Daya, S. Haider, N. Talukdar, D. A. Clausi, and A. Wong,"Spatio-Temporal Saliency Detection Using Abstracted Fully-Connected Graphical Models", International Conference on Image Processing, September, 2016. Details

Scharfenberger, C.A. JainA. Wong, and P. Fieguth, "Image saliency detection via multi-scale statistical non-redundancy modelingpdf", IEEE Conference on Image Processing, 2014. Details