Abstract

Diffusion-based salient region detection methods have gained great popularity. In most diffusion-based methods, the saliency values are ranked on 2-layer neighborhood graph by connecting each node to its neighboring nodes and the nodes sharing common boundaries with its neighboring nodes. However, only considering the local relevance between neighbors, the salient region may be heterogeneous and even wrongly suppressed, especially when the features of salient object are diverse. In order to address the issue, we present an effective saliency detection method using diffusing process on the graph with nonlocal connections. First, a saliency-biased Gaussian model is used to refine the saliency map based on the compactness cue, and then, the saliency information of compactness is diffused on 2-layer sparse graph with nonlocal connections. Second, we obtain the contrast of each superpixel by restricting the reference region to the background. Similarly, a saliency-biased Gaussian refinement model is generated and the saliency information based on the uniqueness cue is propagated on the 2-layer sparse graph. We linearly integrate the initial saliency maps based on the compactness and uniqueness cues due to the complementarity to each other. Finally, to obtain a highlighted and homogeneous saliency map, a single-layer updating and multi-layer integrating scheme is presented. Comprehensive experiments on four benchmark datasets demonstrate that the proposed method performs better in terms of various evaluation metrics.

Highlights

  • Saliency detection, which aims to find the most noteworthy region in a scene, is becoming increasingly important, especially when the amount of image is explosively increasing in the age of Big

  • To prove the effectiveness of our proposed method, we evaluated the performance of the saliency detection methods using three popular evaluation criterions: the average precision-recall curve, F-Measure, and mean absolute error (MAE)

  • It can be converted to a binary mask M and Precision with Recall can be computed by comparing M with ground-truth GT: Precision =

Read more

Summary

Introduction

It has been effectively applied in many computer vision tasks, such as image segmentation [1], object detection and recognition [2,3], and image compression [4]. Many saliency detection methods have been proposed. These models can generally be categorized into top-down and bottom-up methods in terms of the mechanisms. Top-down methods [5,6] are task-driven which generally require supervised learning and need to exploit high-level human perceptual knowledge. Bottom-up methods [7,8,9,10,11,12,13,14,15] are data-driven and usually exploit low-level cues, such as features, colors, and spatial distances to construct saliency maps. Most bottom-up methods adopt compactness [8,9,16], uniqueness [7,13,14,15], and background cues [10,11,12]

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call