Mixed-Supervised Scene Text Detection With Expectation-Maximization Algorithm

Scene text detection is an important and challenging task in computer vision. For detecting arbitrarily-shaped texts, most existing methods require heavy data labeling efforts to produce polygon-level text region labels for supervised training. In order to reduce the cost in data labeling, we study...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on image processing Vol. 31; pp. 5513 - 5528
Main Authors: Zhao, Mengbiao, Feng, Wei, Yin, Fei, Zhang, Xu-Yao, Liu, Cheng-Lin
Format: Journal Article
Language:English
Published: New York IEEE 2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Annotations
Benchmark testing
Benchmarks
Computer vision
Costs
Data models
Datasets
Detectors
expectation-maximization algorithm
Labeling
Labels
Maximization
Mixed-supervised learning
Optimization
Polygons
scene text detection
Supervised learning
Training
weak supervision forms
ISSN:1057-7149, 1941-0042, 1941-0042
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Scene text detection is an important and challenging task in computer vision. For detecting arbitrarily-shaped texts, most existing methods require heavy data labeling efforts to produce polygon-level text region labels for supervised training. In order to reduce the cost in data labeling, we study mixed-supervised arbitrarily-shaped text detection by combining various weak supervision forms (e.g., image-level tags, coarse, loose and tight bounding boxes), which are far easier to annotate. Whereas the existing weakly-supervised learning methods (such as multiple instance learning) do not promote full object coverage, to approximate the performance of fully-supervised detection, we propose an Expectation-Maximization (EM) based mixed-supervised learning framework to train scene text detector using only a small amount of polygon-level annotated data combined with a large amount of weakly annotated data. The polygon-level labels are treated as latent variables and recovered from the weak labels by the EM algorithm. A new contour-based scene text detector is also proposed to facilitate the use of weak labels in our mixed-supervised learning framework. Extensive experiments on six scene text benchmarks show that (1) using only 10% strongly annotated data and 90% weakly annotated data, our method yields comparable performance to that of fully supervised methods, (2) with 100% strongly annotated data, our method achieves state-of-the-art performance on five scene text benchmarks (CTW1500, Total-Text, ICDAR-ArT, MSRA-TD500, and C-SVT), and competitive results on the ICDAR2015 Dataset. We will make our weakly annotated datasets publicly available.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1057-7149
1941-0042
1941-0042
DOI:10.1109/TIP.2022.3197987