Learning AND-OR Templates for Object Recognition and Detection
- 1 February 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 35 (9), 2189-2205
- https://doi.org/10.1109/tpami.2013.35
Abstract
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.This publication has 21 references indexed in Scilit:
- Learning Hybrid Image Templates (HIT) by Information ProjectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Articulated pose estimation with flexible mixtures-of-partsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- Learning Active Basis Model for Object Detection and RecognitionInternational Journal of Computer Vision, 2009
- Discriminative structure learning of hierarchical representations for object detectionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- Unsupervised Learning of Probabilistic Grammar-Markov Models for Object CategoriesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2008
- Unsupervised Category Modeling, Recognition, and Segmentation in ImagesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2008
- Describing Visual Scenes Using Transformed Objects and PartsInternational Journal of Computer Vision, 2007
- Robust Object Recognition with Cortex-Like MechanismsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2007
- Inducing features of random fieldsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978