TY - JOUR
T1 - Adaptive Irregular Graph Construction-Based Salient Object Detection
AU - Zhou, Yuan
AU - Zhang, Tianhao
AU - Huo, Shuwei
AU - Hou, Chunping
AU - Kung, Sun Yuan
N1 - Funding Information:
Manuscript received October 6, 2018; revised November 24, 2018 and January 19, 2019; accepted February 24, 2019. Date of publication March 20, 2019; date of current version June 4, 2020. This work was supported in part by the National Natural Science Foundation of China under Grant 61571326 and Grant 61520106002 and in part by the Natural Science Foundation of Tianjin under Grant 16JCQNJC00900. This paper was recommended by Associate Editor Y. Zhang. (Corresponding author: Yuan Zhou.) Y. Zhou, T. Zhang, S. Huo, and C. Hou are with the School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China (e-mail: zhouyuan@tju.edu.cn; zhang_th@tju.edu.cn; huosw@tju.edu.cn; hcp@tju.edu.cn).
Publisher Copyright:
© 1991-2012 IEEE.
PY - 2020/6
Y1 - 2020/6
N2 - Saliency detection represents a vital pre-processing stage of computer vision. Most existing propagation-based salient object detection methods construct a k-regular graph for saliency propagation. Applying a regular graph to a vast smooth region is potentially prone to unnecessary or prolonged propagation errors, leading to the excessive highlighting of the background regions. To mitigate such problems, we substitute the conventional k-regular graph with an adaptive irregular graph for saliency value propagation, thereby avoiding unnecessary iterations over a vast smooth region. We first perform a clustering analysis based on the smoothness, color, and other features of regions. The new graph boosts an adaptive link density by considering the clustering result. In addition, we propose a seeding strategy for the propagation. Based on our experimental studies of six major benchmark datasets, our method performed favorably against the other state-of-the-art methods, both quantitatively and qualitatively.
AB - Saliency detection represents a vital pre-processing stage of computer vision. Most existing propagation-based salient object detection methods construct a k-regular graph for saliency propagation. Applying a regular graph to a vast smooth region is potentially prone to unnecessary or prolonged propagation errors, leading to the excessive highlighting of the background regions. To mitigate such problems, we substitute the conventional k-regular graph with an adaptive irregular graph for saliency value propagation, thereby avoiding unnecessary iterations over a vast smooth region. We first perform a clustering analysis based on the smoothness, color, and other features of regions. The new graph boosts an adaptive link density by considering the clustering result. In addition, we propose a seeding strategy for the propagation. Based on our experimental studies of six major benchmark datasets, our method performed favorably against the other state-of-the-art methods, both quantitatively and qualitatively.
KW - Salient object detection
KW - adaptive irregular graph
KW - label propagation
UR - http://www.scopus.com/inward/record.url?scp=85086309774&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85086309774&partnerID=8YFLogxK
U2 - 10.1109/TCSVT.2019.2904463
DO - 10.1109/TCSVT.2019.2904463
M3 - Article
AN - SCOPUS:85086309774
SN - 1051-8215
VL - 30
SP - 1569
EP - 1582
JO - IEEE Transactions on Circuits and Systems for Video Technology
JF - IEEE Transactions on Circuits and Systems for Video Technology
IS - 6
M1 - 8671726
ER -