Skip to Main content Skip to Navigation
New interface
Conference papers

Superpixel-based saliency detection

Zhi Liu 1, 2 Olivier Le Meur 2 Shuhua Luo 1 
Abstract : In this paper, we propose an effective superpixel-based saliency model. First, the original image is simplified by performing superpixel segmentation and adaptive color quantization. On the basis of superpixel representation, inter-superpixel similarity measures are then calculated based on difference of histograms and spatial distance between each pair of superpixels. For each superpixel, its global contrast measure and spatial sparsity measure are evaluated, and refined with the integration of intersuperpixel similarity measures to finally generate the superpixel-level saliency map. Experimental results on a dataset containing 1,000 test images with ground truths demonstrate that the proposed saliency model outperforms state-of-the-art saliency models.
Document type :
Conference papers
Complete list of metadata
Contributor : Olivier Le Meur Connect in order to contact the contributor
Submitted on : Thursday, October 24, 2013 - 8:57:31 AM
Last modification on : Thursday, January 20, 2022 - 4:20:10 PM
Long-term archiving on: : Monday, January 27, 2014 - 12:10:39 PM


  • HAL Id : hal-00876184, version 1


Zhi Liu, Olivier Le Meur, Shuhua Luo. Superpixel-based saliency detection. WIAMIS - 14th International Workshop on Image and Audio Analysis for Multimedia Interactive Services, Jul 2013, Paris, France. ⟨hal-00876184⟩



Record views


Files downloads