TY - JOUR
T1 - Discovering thematic objects in image collections and videos
AU - Yuan, Junsong
AU - Zhao, Gangqiang
AU - Fu, Yun
AU - Li, Zhu
AU - Katsaggelos, Aggelos K.
AU - Wu, Ying
N1 - Funding Information:
Manuscript received August 05, 2010; revised February 06, 2011, July 18, 2011 and December 03, 2011; accepted December 03, 2011. Date of publication December 26, 2011; date of current version March 21, 2012. This work was supported in part by the Nanyang Assistant Professorship (SUG M5804001) to Dr. J. Yuan. Dr. Y. Fu was supported in part by Futurewei (Huawei) Technologies Inc. and the Intelligence Community Postdoctoral Research Fellowship under Award 2011-11071400006. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Xuelong Li.
PY - 2012/4
Y1 - 2012/4
N2 - Given a collection of images or a short video sequence, we define a thematic object as the key object that frequently appears and is the representative of the visual contents. Successful discovery of the thematic object is helpful for object search and tagging, video summarization and understanding, etc. However, this task is challenging because 1) there lacks a priori knowledge of the thematic objects, such as their shapes, scales, locations, and times of re-occurrences, and 2) the thematic object of interest can be under severe variations in appearances due to viewpoint and lighting condition changes, scale variations, etc. Instead of using a top-down generative model to discover thematic visual patterns, we propose a novel bottom-up approach to gradually prune uncommon local visual primitives and recover the thematic objects. A multilayer candidate pruning procedure is designed to accelerate the image data mining process. Our solution can efficiently locate thematic objects of various sizes and can tolerate large appearance variations of the same thematic object. Experiments on challenging image and video data sets and comparisons with existing methods validate the effectiveness of our method.
AB - Given a collection of images or a short video sequence, we define a thematic object as the key object that frequently appears and is the representative of the visual contents. Successful discovery of the thematic object is helpful for object search and tagging, video summarization and understanding, etc. However, this task is challenging because 1) there lacks a priori knowledge of the thematic objects, such as their shapes, scales, locations, and times of re-occurrences, and 2) the thematic object of interest can be under severe variations in appearances due to viewpoint and lighting condition changes, scale variations, etc. Instead of using a top-down generative model to discover thematic visual patterns, we propose a novel bottom-up approach to gradually prune uncommon local visual primitives and recover the thematic objects. A multilayer candidate pruning procedure is designed to accelerate the image data mining process. Our solution can efficiently locate thematic objects of various sizes and can tolerate large appearance variations of the same thematic object. Experiments on challenging image and video data sets and comparisons with existing methods validate the effectiveness of our method.
KW - Image data mining
KW - thematic object discovery
UR - http://www.scopus.com/inward/record.url?scp=84859024517&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84859024517&partnerID=8YFLogxK
U2 - 10.1109/TIP.2011.2181952
DO - 10.1109/TIP.2011.2181952
M3 - Article
C2 - 22207639
AN - SCOPUS:84859024517
VL - 21
SP - 2207
EP - 2219
JO - IEEE Transactions on Image Processing
JF - IEEE Transactions on Image Processing
SN - 1057-7149
IS - 4
M1 - 6112717
ER -