Deep features for text spotting

Conference item

The goal of this work is text spotting in natural images. This is divided into two sequential tasks: detecting words regions in the image, and recognizing the words within these regions. We make the following contributions: first, we develop a Convolutional Neural Network (CNN) classifier that can be used for both tasks. The CNN has a novel architecture that enables efficient feature sharing (by using a number of layers in common) for text detection, character case-sensitive and insensitive classification, and bigram classification. It exceeds the state-of-the-art performance for all of these. Second, we make a number of technical changes over the traditional CNN architectures, including no downsampling for a per-pixel sliding window, and multi-mode learning with a mixture of linear models (maxout). Third, we have a method of automated data mining of Flickr, that generates word and character level annotations. Finally, these components are used together to form an end-to-end, state-of-the-art text spotting system. We evaluate the text-spotting system on two standard benchmarks, the ICDAR Robust Reading data set and the Street View Text data set, and demonstrate improvements over the state-of-the-art on multiple measures. © 2014 Springer International Publishing.

Authors: Jaderberg, M; Vedaldi, A; Zisserman, A

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Springer Nature
• Host title: Computer Vision -- ECCV 2014 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part IV
• Pages: 512-528
• Series: Lecture Notes in Computer Science
• Series number: 8692
• Publication date: 2014-08-14
• Acceptance date: 2014-06-17
• Event title: 13th European Conference, Computer Vision - ECCV 2014
• Event location: Zurich, Switzerland
• Event start date: 2014-09-06
• Event end date: 2014-09-12
• DOI: 10.1007/978-3-319-10593-2_34
• EISSN: 1611-3349
• ISSN: 0302-9743
• EISBN: 9783319105932
• ISBN: 9783319105925
• Language: English