Learning AND-OR Templates for Object Recognition and Detection

By Zhangzhang Si, Song-Chun Zhu. 2013

127 citations recorded by the source corpus.

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.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Machine Learning and Reasoning (2019, background): “Besides formal logical knowledge, hierarchical Boolean expressions can be learned from images for object detection and recognition [44].”
Cross-View Action Modeling, Learning, and Recognition (2014, methodology): “We call this model multiview spatio-temporal ANDOR graph model (MST-AOG), inspired by the expressive power of AND-OR graphs in object modeling [18].”
Mining And-Or Graphs for Graph Matching and Object Discovery (2015, methodology): “For each OR node s, we apply a standard inference strategy for OR nodes [25, 31], i.”
Modeling Occlusion by Discriminative AND-OR Structures (2013, methodology): “5 using a modified graph compression algorithm similar to [27].”
Learning Multi-Way Relations via Tensor Decomposition With Neural Networks (2018, background): “Si et al. learned compositional models of objects in images (Si and Zhu 2013).”
Mining Object Parts from CNNs via Active Question-Answering (2017, methodology): “AOG structure is suitable for representing semantic hierarchy of objects [50, 29], and [46] used an AOG to represent the CNN.”
Unsupervised learning of event AND-OR grammar and semantics from video (2011, methodology): “We refer to [9] for a detailed derivation of the block pursuit procedure.”
Teaching Compositionality to CNNs (2017, background): “The first flavor focuses on the creation of hierarchical feature representations by means of statistical modeling [8, 27, 48, 47], reusable deformable parts [49, 29], or compositional graph structures [36, 45].”
Interactively Transferring CNN Patterns for Part Localization (2017, background): “And-Or representation: In many studies, people used AOGs to represent the hierarchical semantic hierarchy of objects or scenes [22, 11].”
Explanation and Justification in Machine Learning : A Survey Or (2017, background): “[Si and Zhu, 2013] learn compositional models of objects in images: an object (e.g., a cat) contains a mandatory set of parts (e.g., ears), but each part can come in many forms (pointed, round...), and each form is represented using a pixel-level generative model.”
From Red Wine to Red Tomato: Composition with Context (2017, background): “The parts-based systems such as Deformable Part Models [15], grammars [25, 44, 59, 62, 67, 69], and AND-OR graphs [54, 61, 66] also rely on compositionality of objects to build recognition systems.”