Name
Abstract | ||
|---|---|---|
This paper presents a method of learning reconfigurable And-Or Tree (AOT) models discriminatively from weakly annotated data for object detection. To explore the appearance and geometry space of latent structures effectively, we first quantize the image lattice using an over complete set of shape primitives, and then organize them into a directed a cyclic And-Or Graph (AOG) by exploiting their compositional relations. We allow overlaps between child nodes when combining them into a parent node, which is equivalent to introducing an appearance Or-node implicitly for the overlapped portion. The learning of an AOT model consists of three components: (i) Unsupervised sub-category learning (i.e., branches of an object Or-node) with the latent structures in AOG being integrated out. (ii) Weakly supervised part configuration learning (i.e., seeking the globally optimal parse trees in AOG for each sub-category). To search the globally optimal parse tree in AOG efficiently, we propose a dynamic programming (DP) algorithm. (iii) Joint appearance and structural parameters training under latent structural SVM framework. In experiments, our method is tested on PASCAL VOC 2007 and 2010 detection benchmarks of 20 object classes and outperforms comparable state-of-the-art methods. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/CVPR.2013.421 | CVPR |
Keywords | Field | DocType |
optimal parse tree,appearance or-node,latent structural svm,structural svm framework,object class,aot model,learning (artificial intelligence),joint appearance,reconfigurable and-or tree models,object detection,unsupervised sub-category learning,structural parameters training,latent structural svm framework,2010 detection benchmarks,geometry space,part-based representation,discriminatively trained and-or tree models,latent structure,weakly annotated data,and-or tree models,globally optimal parse trees,directed graphs,dynamic programming algorithm,shape primitives,image lattice,directed a cyclic and-or graph,dynamic programming,pascal voc 2007,and-or graph,object or-node,space exploration,support vector machines,learning artificial intelligence,shape,lattices | Parse tree,Lattice (order),Computer science,Artificial intelligence,Computer vision,Dynamic programming,Object detection,Pattern recognition,Support vector machine,Directed graph,Parsing,Machine learning,And–or tree | Conference |
ISSN | Citations | PageRank |
1063-6919 | 30 | 0.93 |
References | Authors | |
18 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xi Song | 1 | 45 | 5.28 |
| Tianfu Wu | 2 | 331 | 26.72 |
| Yunde Jia | 3 | 526 | 26.24 |
| Song-Chun Zhu | 4 | 6580 | 741.75 |