Using Contextual Words for Object Detection

Jun Guo; Chang Ren Zhu

doi:10.4028/www.scientific.net/AMM.190-191.227

Paper Titles

The Solution of the Dynamics Prediction of Aircraft State
p.209

The TCA of Hypoid Gears Based on MATLAB
p.213

The Uncertain Programming of the Closed-Loop Supply Chain with Remanufacturing System
p.218

Three Dimensional Numerical Analysis of Plasma Spraying Using Supersonic Plasma Gun
p.222

Using Contextual Words for Object Detection
p.227

Virtual Prototype and Modal Analysis of Stator System of Large Turbo-Generator
p.232

Web Roaming of 3D Taiwan Island Based on Skyline Prototype System
p.237

A New Approach to Battery Capacity Prediction Based on Hybrid ARMA and ANN Model
p.241

Finite-Element Analyze of D-Post on Numeric Controlled Gear Hobbing Cooperative Working Machine Based on ANSYS
p.245

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 190-191Using Contextual Words for Object Detection

Using Contextual Words for Object Detection

Abstract:

Recent work has shown that classification can benefit saliently from the proper use of contextual cues. In this paper, we present a novel framework by using contextual information to improve object detection. After having detected objects and clustered image regions respectively, we make the statistic spatial relationships between candidate objects and sub-regions of the image, which are composed into a vector of features (termed contextual words). Then, the vectors of all candidates are used to train the probabilistic Latent Semantic Analysis (pLSA) model. Finally, SVM classifier is employed in object classification. Furthermore, a new direct graphical model by extending pLSA is proposed to introduce more useful information from previous detection process. The experimental results based on a remote-sensing image dataset show that our performances are quite promising.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 190-191)

Pages:

227-231

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.190-191.227

Citation:

Cite this paper

Online since:

July 2012

Authors:

Jun Guo, Chang Ren Zhu

Keywords:

Contextual Word, Object Detection, Probabilistic Latent Semantic Analysis

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] A. Bosch, A. Zisserman, et al., Scene classification using a hybrid generative/discriminative approach, IEEE Transactions on PAMI 30(4) (2008) 712-727.

DOI: 10.1109/tpami.2007.70716

Google Scholar

[2] G. Heitz, D. Koller, Learning Spatial Context: Using Stuff to Find Things, ECCV, Part I, LNCS 5302, (2008) 30-43.

DOI: 10.1007/978-3-540-88682-2_4

Google Scholar

[3] H. Zhou, R. Wang, and C. Wang, A novel extended local-binary-pattern operator for texture analysis, Information Sciences 178 (2008) 4314-4325.

DOI: 10.1016/j.ins.2008.07.015

Google Scholar

[4] A. Torralba, Contextual priming for object detection. IJCV 53(2) (2003).

Google Scholar

[5] X. Ren, J. Malik, Learning a classification model for segmentation, ICCV (2003).

Google Scholar

[6] T. Hofmann, Unsupervised learning by probabilistic latent semantic analysis, Machine Learning 42 (1–2) (2001) 177-196.

Google Scholar

[7] N. Ueda, R. Nakano, Deterministic annealing EM algorithm, Neural Networks 11(2) (1988) 271-282.

DOI: 10.1016/s0893-6080(97)00133-0

Google Scholar

[8] V. Vapnik, Statistical Learning Theory, Wiley, New York, (1998).

Google Scholar