Ranking Saliency

Lihe Zhang, Chuanr Yang, Huchuan Lu, Xiang Ruan, Ming Hsuan Yang

Research output: Contribution to journalArticle

50 Citations (Scopus)

Abstract

Most existing bottom-up algorithms measure the foreground saliency of a pixel or region based on its contrast within a local context or the entire image, whereas a few methods focus on segmenting out background regions and thereby salient objects. Instead of only considering the contrast between salient objects and their surrounding regions, we consider both foreground and background cues in this work. We rank the similarity of image elements with foreground or background cues via graph-based manifold ranking. The saliency of image elements is defined based on their relevances to the given seeds or queries. We represent an image as a multi-scale graph with fine superpixels and coarse regions as nodes. These nodes are ranked based on the similarity to background and foreground queries using affinity matrices. Saliency detection is carried out in a cascade scheme to extract background regions and foreground salient objects efficiently. Experimental results demonstrate the proposed method performs well against the state-of-the-art methods in terms of accuracy and speed. We also propose a new benchmark dataset containing 5,168 images for large-scale performance evaluation of saliency detection methods.

Original languageEnglish
Article number7567535
Pages (from-to)1892-1904
Number of pages13
JournalIEEE transactions on pattern analysis and machine intelligence
Volume39
Issue number9
DOIs
Publication statusPublished - 2017 Sep 1

Fingerprint

Saliency
Seed
Ranking
Pixels
Query
Graph in graph theory
Vertex of a graph
Bottom-up
Cascade
Affine transformation
Performance Evaluation
Pixel
Background
Entire
Benchmark
Experimental Results
Demonstrate
Object

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

Zhang, Lihe ; Yang, Chuanr ; Lu, Huchuan ; Ruan, Xiang ; Yang, Ming Hsuan. / Ranking Saliency. In: IEEE transactions on pattern analysis and machine intelligence. 2017 ; Vol. 39, No. 9. pp. 1892-1904.
@article{726f085ba5c54ae9afb9bc0f10ed9eb1,
title = "Ranking Saliency",
abstract = "Most existing bottom-up algorithms measure the foreground saliency of a pixel or region based on its contrast within a local context or the entire image, whereas a few methods focus on segmenting out background regions and thereby salient objects. Instead of only considering the contrast between salient objects and their surrounding regions, we consider both foreground and background cues in this work. We rank the similarity of image elements with foreground or background cues via graph-based manifold ranking. The saliency of image elements is defined based on their relevances to the given seeds or queries. We represent an image as a multi-scale graph with fine superpixels and coarse regions as nodes. These nodes are ranked based on the similarity to background and foreground queries using affinity matrices. Saliency detection is carried out in a cascade scheme to extract background regions and foreground salient objects efficiently. Experimental results demonstrate the proposed method performs well against the state-of-the-art methods in terms of accuracy and speed. We also propose a new benchmark dataset containing 5,168 images for large-scale performance evaluation of saliency detection methods.",
author = "Lihe Zhang and Chuanr Yang and Huchuan Lu and Xiang Ruan and Yang, {Ming Hsuan}",
year = "2017",
month = "9",
day = "1",
doi = "10.1109/TPAMI.2016.2609426",
language = "English",
volume = "39",
pages = "1892--1904",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "9",

}

Ranking Saliency. / Zhang, Lihe; Yang, Chuanr; Lu, Huchuan; Ruan, Xiang; Yang, Ming Hsuan.

In: IEEE transactions on pattern analysis and machine intelligence, Vol. 39, No. 9, 7567535, 01.09.2017, p. 1892-1904.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Ranking Saliency

AU - Zhang, Lihe

AU - Yang, Chuanr

AU - Lu, Huchuan

AU - Ruan, Xiang

AU - Yang, Ming Hsuan

PY - 2017/9/1

Y1 - 2017/9/1

N2 - Most existing bottom-up algorithms measure the foreground saliency of a pixel or region based on its contrast within a local context or the entire image, whereas a few methods focus on segmenting out background regions and thereby salient objects. Instead of only considering the contrast between salient objects and their surrounding regions, we consider both foreground and background cues in this work. We rank the similarity of image elements with foreground or background cues via graph-based manifold ranking. The saliency of image elements is defined based on their relevances to the given seeds or queries. We represent an image as a multi-scale graph with fine superpixels and coarse regions as nodes. These nodes are ranked based on the similarity to background and foreground queries using affinity matrices. Saliency detection is carried out in a cascade scheme to extract background regions and foreground salient objects efficiently. Experimental results demonstrate the proposed method performs well against the state-of-the-art methods in terms of accuracy and speed. We also propose a new benchmark dataset containing 5,168 images for large-scale performance evaluation of saliency detection methods.

AB - Most existing bottom-up algorithms measure the foreground saliency of a pixel or region based on its contrast within a local context or the entire image, whereas a few methods focus on segmenting out background regions and thereby salient objects. Instead of only considering the contrast between salient objects and their surrounding regions, we consider both foreground and background cues in this work. We rank the similarity of image elements with foreground or background cues via graph-based manifold ranking. The saliency of image elements is defined based on their relevances to the given seeds or queries. We represent an image as a multi-scale graph with fine superpixels and coarse regions as nodes. These nodes are ranked based on the similarity to background and foreground queries using affinity matrices. Saliency detection is carried out in a cascade scheme to extract background regions and foreground salient objects efficiently. Experimental results demonstrate the proposed method performs well against the state-of-the-art methods in terms of accuracy and speed. We also propose a new benchmark dataset containing 5,168 images for large-scale performance evaluation of saliency detection methods.

UR - http://www.scopus.com/inward/record.url?scp=85029380268&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029380268&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2016.2609426

DO - 10.1109/TPAMI.2016.2609426

M3 - Article

C2 - 28113655

AN - SCOPUS:85029380268

VL - 39

SP - 1892

EP - 1904

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 9

M1 - 7567535

ER -