Abstract | ||
---|---|---|
In this work, a novel salient object detection method is proposed based on the saliency driven clustering. To capture visual patterns of an image, the color contrast prior and boundary prior are utilized to generate the image clusters automatically. Then, a simple operation like regional saliency computation is applied to refine the saliency maps generated by two priors. The final saliency map are obtained by combining the refined contrast prior saliency and boundary prior saliency. Extensive experiments show that our proposed model achieves better performance on salient region detection against the state-of-the-art methods. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/ICASSP.2014.6854629 | ICASSP |
Keywords | Field | DocType |
image coding,color contrast prior,salient region detection,regional saliency computation,boundary prior,object detection,salient object detection,visual image patterns,saliency driven clustering,computational modeling,visualization,computer vision,pattern recognition | Color contrast,Computer vision,Salient object detection,Kadir–Brady saliency detector,Pattern recognition,Salience (neuroscience),Computer science,Artificial intelligence,Prior probability,Cluster analysis,Computation,Salient | Conference |
ISSN | Citations | PageRank |
1520-6149 | 1 | 0.34 |
References | Authors | |
11 | 5 |