Regular Articles

Spatial encoding of visual words for image classification

[+] Author Affiliations
Dong Liu, Shengsheng Wang

Jilin University, College of Computer Science and Technology, Changchun 130012, China

Jilin University, Key Laboratory of Symbolic Computation and Knowledge Engineering of the Ministry of Education, Changchun 130012, China

Fatih Porikli

Australian National University, Research School of Engineering, RSISE Building, Canberra 2601, ACT, Australia

Data61, 7 London Circuit, Canberra 2601, ACT, Australia

J. Electron. Imaging. 25(3), 033008 (May 31, 2016). doi:10.1117/1.JEI.25.3.033008
History: Received January 7, 2016; Accepted May 4, 2016
Text Size: A A A

Abstract.  Appearance-based bag-of-visual words (BoVW) models are employed to represent the frequency of a vocabulary of local features in an image. Due to their versatility, they are widely popular, although they ignore the underlying spatial context and relationships among the features. Here, we present a unified representation that enhances BoVWs with explicit local and global structure models. Three aspects of our method should be noted in comparison to the previous approaches. First, we use a local structure feature that encodes the spatial attributes between a pair of points in a discriminative fashion using class-label information. We introduce a bag-of-structural words (BoSW) model for the given image set and describe each image with this model on its coarsely sampled relevant keypoints. We then combine the codebook histograms of BoVW and BoSW to train a classifier. Rigorous experimental evaluations on four benchmark data sets demonstrate that the unified representation outperforms the conventional models and compares favorably to more sophisticated scene classification techniques.

Figures in this Article
© 2016 SPIE and IS&T

Citation

Dong Liu ; Shengsheng Wang and Fatih Porikli
"Spatial encoding of visual words for image classification", J. Electron. Imaging. 25(3), 033008 (May 31, 2016). ; http://dx.doi.org/10.1117/1.JEI.25.3.033008


Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Related Book Chapters

Topic Collections

PubMed Articles
Advertisement
  • Don't have an account?
  • Subscribe to the SPIE Digital Library
  • Create a FREE account to sign up for Digital Library content alerts and gain access to institutional subscriptions remotely.
Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).
Access This Proceeding
Sign in or Create a personal account to Buy this article ($15 for members, $18 for non-members).
Access This Chapter

Access to SPIE eBooks is limited to subscribing institutions and is not available as part of a personal subscription. Print or electronic versions of individual SPIE books may be purchased via SPIE.org.