Natural Scene Character Recognition Without Dependency On Specific Features

Keywords

Feature independence; Holistic character recognition; Natural scene text recognition; Rank-1 decomposition; Tensors

Abstract

Current methods in scene character recognition heavily rely on discriminative power of local features, such as HoG, SIFT, Shape Contexts (SC), Geometric Blur (GB), etc. One of the problems with this approach is that the local features are rasterized in an ad hoc manner into a single vector perturbing thus spatial correlations that carry crucial information. To eliminate this feature dependency and associated problems, we propose a holistic solution as follows: For each character to be recognized, we stack a set of training images to form a 3-mode tensor. Each training tensor is then decomposed into a linear superposition of 'k' rank-1 matrices, whereby the rank-1 matrices form a basis, spanning solution subspace of the character class. For a test image to be classified, we obtain projections onto the pre-computed rank-1 bases of each class, and recognize it as the class for which inner-product of mixing vectors is maximized. We use challenging natural scene character datasets, namely Chars74K, ICDAR2003, and SVT-CHAR. We achieve results better than several baseline methods based on local features (e.g. HoG) and show leave-random-one-out-cross validation yield even better recognition performance, justifying thus our intuition of the importance of feature-independency and preservation of spatial correlations in recognition.

Publication Date

1-1-2015

Publication Title

VISAPP 2015 - 10th International Conference on Computer Vision Theory and Applications; VISIGRAPP, Proceedings

Volume

2

Number of Pages

368-376

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

DOI Link

https://doi.org/10.5220/0005305603680376

Socpus ID

84939518799 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/84939518799

This document is currently not available here.

Share

COinS