TY - JOUR
T1 - Unsupervised Primary Object Discovery in Videos Based on Evolutionary Primary Object Modeling with Reliable Object Proposals
AU - Koh, Yeong Jun
AU - Kim, Chang-Su
N1 - Funding Information:
Manuscript received December 24, 2016; revised April 27, 2017 and June 20, 2017; accepted July 27, 2017. Date of publication August 4, 2017; date of current version August 21, 2017. This work was supported in part by the National Research Foundation of Korea (NRF) grant funded by the Korean government under Grant NRF-2015R1A2A1A10055037, and in part by the Agency for Defense Development and Defense Acquisition Program Administration, South Korea under Grant UC160016FD. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Khan M. Iftekharuddin. (Corresponding author: Chang-Su Kim.) The authors are with the School of Electrical Engineering, Korea University, Seoul 02841, South Korea (e-mail: yjkoh@mcl.korea.ac.kr; changsukim@korea.ac.kr).
Publisher Copyright:
© 1992-2012 IEEE.
PY - 2017/11
Y1 - 2017/11
N2 - A novel primary object discovery (POD) algorithm, which uses reliable object proposals while exploiting the recurrence property of a primary object in a video sequence, is proposed in this paper. First, we generate both color-based and motion-based object proposals in each frame, and extract the feature of each proposal using the random walk with restart simulation. Next, we estimate the foreground confidence for each proposal to remove unreliable proposals. By superposing the features of the remaining reliable proposals, we construct the primary object models. To this end, we develop the evolutionary primary object modeling technique, which exploits the recurrence property of the primary object. Then, using the primary object models, we choose the main proposal in each frame and find the location of the primary object by merging the main proposal with candidate proposals selectively. Finally, we refine the discovered bounding boxes by exploiting temporal correlations of the recurring primary object. Extensive experimental results demonstrate that the proposed POD algorithm significantly outperforms conventional algorithms.
AB - A novel primary object discovery (POD) algorithm, which uses reliable object proposals while exploiting the recurrence property of a primary object in a video sequence, is proposed in this paper. First, we generate both color-based and motion-based object proposals in each frame, and extract the feature of each proposal using the random walk with restart simulation. Next, we estimate the foreground confidence for each proposal to remove unreliable proposals. By superposing the features of the remaining reliable proposals, we construct the primary object models. To this end, we develop the evolutionary primary object modeling technique, which exploits the recurrence property of the primary object. Then, using the primary object models, we choose the main proposal in each frame and find the location of the primary object by merging the main proposal with candidate proposals selectively. Finally, we refine the discovered bounding boxes by exploiting temporal correlations of the recurring primary object. Extensive experimental results demonstrate that the proposed POD algorithm significantly outperforms conventional algorithms.
KW - Primary object discovery
KW - object proposal
KW - recurrence property
KW - video object segmentation
UR - http://www.scopus.com/inward/record.url?scp=85029009755&partnerID=8YFLogxK
U2 - 10.1109/TIP.2017.2736418
DO - 10.1109/TIP.2017.2736418
M3 - Article
C2 - 28792896
AN - SCOPUS:85029009755
SN - 1057-7149
VL - 26
SP - 5203
EP - 5216
JO - IEEE Transactions on Image Processing
JF - IEEE Transactions on Image Processing
IS - 11
M1 - 8002643
ER -