This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Learning AND-OR Templates for Object Recognition and Detection
Sept. 2013 (vol. 35 no. 9)
pp. 2189-2205
Zhangzhang Si, Dept. of Stat., Univ. of California, Los Angeles, Los Angeles, CA, USA
Song-Chun Zhu, Dept. of Stat., Univ. of California, Los Angeles, Los Angeles, CA, USA
This paper presents a framework for unsupervised learning of a hierarchical reconfigurable image template - the AND-OR Template (AOT) for visual objects. The AOT includes: 1) hierarchical composition as "AND" nodes, 2) deformation and articulation of parts as geometric "OR" nodes, and 3) multiple ways of composition as structural "OR" nodes. The terminal nodes are hybrid image templates (HIT) [17] that are fully generative to the pixels. We show that both the structures and parameters of the AOT model can be learned in an unsupervised way from images using an information projection principle. The learning algorithm consists of two steps: 1) a recursive block pursuit procedure to learn the hierarchical dictionary of primitives, parts, and objects, and 2) a graph compression procedure to minimize model structure for better generalizability. We investigate the factors that influence how well the learning algorithm can identify the underlying AOT. And we propose a number of ways to evaluate the performance of the learned AOTs through both synthesized examples and real-world images. Our model advances the state of the art for object detection by improving the accuracy of template matching.
Index Terms:
unsupervised learning,generalisation (artificial intelligence),image matching,object detection,object recognition,template matching,AND-OR template learning,object recognition,object detection,unsupervised learning,hierarchical reconfigurable image template,visual object,hierarchical composition,part deformation,part articulation,information projection principle,recursive block pursuit procedure,graph compression procedure,generalizability,Training,Histograms,Image color analysis,Unsupervised learning,Visualization,Animals,Face,information projection,Deformable templates,object recognition,image grammar
Citation:
Zhangzhang Si, Song-Chun Zhu, "Learning AND-OR Templates for Object Recognition and Detection," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35, no. 9, pp. 2189-2205, Sept. 2013, doi:10.1109/TPAMI.2013.35
Usage of this product signifies your acceptance of the Terms of Use.