Skip navigation
Please use this identifier to cite or link to this item: http://repository.iitr.ac.in/handle/123456789/21809
Title: Quadbox: Quadrilateral Bounding Box Based Scene Text Detection Using Vector Regression
Authors: Keserwani P.
Dhankhar A.
Saini R.
Pratim Roy, Partha
Published in: IEEE Access
Abstract: Scene text appears with a wide range of sizes and arbitrary orientations. For detecting such text in the scene image, the quadrilateral bounding boxes provide a much tight bounding box compared to the rotated rectangle. In this work, a vector regression method has been proposed for text detection in the wild to generate a quadrilateral bounding box. The bounding box prediction using direct regression requires predicting the vectors from each position inside the quadrilateral. It needs to predict four-vectors, and each varies drastically in its length and orientation. It makes the vector prediction a difficult problem. To overcome this, we have proposed a centroid-centric vector regression by utilizing the geometry of quadrilateral. In this work, we have added the philosophy of indirect regression to direct regression by shifting all points within the quadrilateral to the centroid and afterward performed vector regression from shifted points. The experimental results show the improvement of the quadrilateral approach over the existing direct regression approach. The proposed method shows good performance on many existing public datasets. The proposed method also demonstrates good results on the unseen dataset without getting trained on it, which validates the approach's generalization ability. © 2013 IEEE.
Citation: IEEE Access, 9: 36802-36818
URI: https://doi.org/10.1109/ACCESS.2021.3063030
http://repository.iitr.ac.in/handle/123456789/21809
Issue Date: 2021
Publisher: Institute of Electrical and Electronics Engineers Inc.
Keywords: centroid of the quadrilateral
direct regression
indirect regression
quadrilateral bounding boxes
Scene text detection
ISSN: 21693536
Author Scopus IDs: 57205562856
57215496563
57190288840
56880478500
Author Affiliations: Keserwani, P., Department of Computer Science and Engineering, IIT Roorkee, Roorkee, 247667, India
Dhankhar, A., Department of Computer Science and Engineering, IIT Roorkee, Roorkee, 247667, India
Saini, R., Department of Computer Science Electrical and Space Engineering, Luleå Tekniska Universitet, Luleå, 971 87, Sweden
Roy, P.P., Department of Computer Science and Engineering, IIT Roorkee, Roorkee, 247667, India
Funding Details: This work was supported by the Vinnova Nationellt Rymddatalabb (Swedish Space Data Lab) Project. The work of Prateek Keserwani was supported by the Visvesvaraya Ph.D. Fellowship.
Corresponding Author: Keserwani, P.; Department of Computer Science and Engineering, India; email: pkeserwani@cs.iitr.ac.in Saini, R.; Department of Computer Science Electrical and Space Engineering, Sweden; email: rajkumar.saini@ltu.se
Appears in Collections:Journal Publications [CS]

Files in This Item:
There are no files associated with this item.
Show full item record


Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.