Image saliency estimation via random walk guided by informativeness and latent signal correlations |
| |
| Affiliation: | 1. State Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, Beijing 100191, China;2. International Research Institute for Multidisciplinary Science, Beihang University, China;3. National Engineering Laboratory for Video Technology, School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China;1. Department of ECE, University of Thessaly, Volos 38221, Greece;2. High Performance Network Lab, Chinese Academy of Sciences, Beijing 100190, China;3. Department of Electrical Engineering, Yale University, New Haven, 06511 CT, USA;4. University of Nebraska-Lincoln, Omaha 68046, USA;1. COSIM Lab., SUP׳COM, Carthage Univ., Cité Technologique des Communications, Tunisia;2. Institut Galilée, L2TI, Université Paris 13, Sorbonne Paris Cité, France;1. National Engineering Laboratory for Video Technology, School of Electronics Engineering and Computer Science, Peking University, China;2. Department of Electronic Engineering, Beijing Institute of Technology, China;3. State Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, China;1. School of Chemical and Environmental Engineering, Qingdao University, Qingdao 266071, China;2. National Marine Environmental Forecasting Center, State Oceanic Administration, Beijing 100081, China |
| |
| Abstract: | Visual saliency is an effective tool for perceptual image processing. In the past decades, many saliency models have been proposed by primarily considering visual cues such as local contrast and global rarity. However, such explicit cues derived only from input stimuli are often insufficient to separate targets from distractors, leading to noisy saliency maps. In fact, the latent cues, especially the latent signal correlations that link visually distinct stimuli (e.g., various parts of a salient target), may also play an important role in saliency estimation. In this paper, we propose a graph-based approach for image saliency estimation by incorporating both explicit visual cues and latent signal correlations. In our approach, the latent correlations between various image patches are first derived according to the statistical prior obtained from 10 million reference images. After that, the informativeness of image patches and their latent correlations are jointly considered to construct a directed graph, on which a random walking process is performed to generate saliency maps that pop-out only the most salient locations. Experimental results show that our approach achieves impressive performances on three public image benchmarks. |
| |
| Keywords: | Image saliency Random walk Informativeness Latent correlation |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
