Detecting oriented text in natural images by linking segments
Publikation: Bidrag til tidsskrift › Konferenceartikel › Forskning › fagfællebedømt
Most state-of-the-art text detection methods are specific to horizontal Latin text and are not fast enough for real-time applications. We introduce Segment Linking (SegLink), an oriented text detection method. The main idea is to decompose text into two locally detectable elements, namely segments and links. A segment is an oriented box covering a part of a word or text line; A link connects two adjacent segments, indicating that they belong to the same word or text line. Both elements are detected densely at multiple scales by an end-to-end trained, fully-convolutional neural network. Final detections are produced by combining segments connected by links. Compared with previous methods, SegLink improves along the dimensions of accuracy, speed, and ease of training. It achieves an f-measure of 75.0% on the standard ICDAR 2015 Incidental (Challenge 4) benchmark, outperforming the previous best by a large margin. It runs at over 20 FPS on 512×512 images. Moreover, without modification, SegLink is able to detect long lines of non-Latin text, such as Chinese.
Originalsprog | Engelsk |
---|---|
Tidsskrift | Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
Sider (fra-til) | 3482-3490 |
Antal sider | 9 |
DOI | |
Status | Udgivet - 6 nov. 2017 |
Eksternt udgivet | Ja |
Begivenhed | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 - Honolulu, USA Varighed: 21 jul. 2017 → 26 jul. 2017 |
Konference
Konference | 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 |
---|---|
Land | USA |
By | Honolulu |
Periode | 21/07/2017 → 26/07/2017 |
Bibliografisk note
Publisher Copyright:
© 2017 IEEE.
ID: 301827309