2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition 2018
DOI: 10.1109/cvpr.2018.00584
|View full text |Cite
|
Sign up to set email alerts
|

AON: Towards Arbitrarily-Oriented Text Recognition

Abstract: Recognizing text from natural images is a hot research topic in computer vision due to its various applications. Despite the enduring research of several decades on optical character recognition (OCR), recognizing texts from natural images is still a challenging task. This is because scene texts are often in irregular (e.g. curved, arbitrarilyoriented or seriously distorted) arrangements, which have not yet been well addressed in the literature. Existing methods on text recognition mainly work with regular (ho… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
200
1

Year Published

2018
2018
2022
2022

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 301 publications
(202 citation statements)
references
References 33 publications
(56 reference statements)
1
200
1
Order By: Relevance
“…In [73], [75], they propose to recognize irregular text by applying attention mechanism on two-dimensional feature maps. Cheng et al [9] propose to encode the input image to four feature sequences of four directions.…”
Section: Scene Text Recognitionmentioning
confidence: 99%
“…In [73], [75], they propose to recognize irregular text by applying attention mechanism on two-dimensional feature maps. Cheng et al [9] propose to encode the input image to four feature sequences of four directions.…”
Section: Scene Text Recognitionmentioning
confidence: 99%
“…Deep learning based methods [20,17,25,48,45,44,7,29,3,8,46,57] have been dominating this area in recent years. Jaderberg et al [20] propose to take scene text recognition as a word classification problem by using a CNN classifier.…”
Section: Text Recognitionmentioning
confidence: 99%
“…There have been a lot of works that focus on dealing with irregular text instances. AON [8] applies sequence recognition * corresponding author. Figure 1: Comparison between ASTER [46] and ScRN (proposed in this paper), shown in (a) and (b) respectively.…”
Section: Introductionmentioning
confidence: 99%
“…According to the results, finally we choose to apply deformable layers in the fourth and fifth convolution. We make a few comparisons with other methods including [1,9,5,6,12,7,24,25,26,27,28,29,30,31,3,32]. All the results are reached without lexicon, which are shown in Table 3.…”
Section: The Impacts Of Location Of Deformable Layersmentioning
confidence: 97%