Robust Text Image Recognition via Adversarial Sequence-to-Sequence Domain Adaptation

Zhang, Yaping; Liang, Shan; Liu, Wenju

doi:10.1109/tip.2021.3066903

Cited by 23 publications

(18 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the popularization and development of Internet technology, image recognition technology began to be applied to all aspects of life [ 1 – 3 ]. Among them, as a common way of communication, freehand sketch has attracted more and more researchers' attention [ 4 ].…”

Section: Introductionmentioning

confidence: 99%

[Retracted] Research on Recognition Effect of DSCN Network Structure in Hand‐Drawn Sketch

2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

With the rapid development of image recognition technology, freehand sketch recognition has attracted more and more attention. How to achieve good recognition effect in the absence of color and texture information is the key to the development of freehand sketch recognition. Traditional nonlearning classical models are highly dependent on manual selection features. To solve this problem, a neural network sketch recognition method based on DSCN structure is proposed in this paper. Firstly, the stroke sequence of the sketch is drawn; then, the feature is extracted according to the stroke sequence combined with neural network, and the extracted image features are used as the input of the model to construct the time relationship between different image features. Through the control experiment on TU-Berlin dataset, the results show that, compared with the traditional nonlearning methods, HOG-SVM, SIFT-Fisher Vector, MKL-SVM, and FV-SP, the recognition accuracy of DSCN network is improved by 15.8%, 10.3%, 6.0%, and 2.9%, respectively. Compared with the classical deep learning model, Alex-Net, the recognition accuracy is improved by 5.6%. The above results show that the DSCN network proposed in this paper has strong ability of feature extraction and nonlinear expression and can effectively improve the recognition accuracy of hand-painted sketches after introducing the stroke order.

show abstract

Section: Introductionmentioning

confidence: 99%

[Retracted] Research on Recognition Effect of DSCN Network Structure in Hand‐Drawn Sketch

2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…To compare our proposed method with existing methods, our experiments' training and evaluation procedures follow the same protocol in [2,22]. In the protocol, we pre-trained the model with synthetic scene text datasets as the source domain and validated on real scene text datasets as the target domain.…”

Section: Methodsmentioning

confidence: 99%

“…GA-DAN generates real scene text images by converting synthetic images but omits domain shift in fine-grained character-level images. Unlike minimizing the domain difference only with the global representation for sequence-to-sequence tasks, an adversarial sequence-tosequence domain adaptation (ASSDA) [22] is proposed to address the issue by the sequential architecture. The authors use global and local alignment to minimize the domain difference to address fine-grained character level domain shift.…”

Section: Related Workmentioning

confidence: 99%

“…The number of fiducial points of TPS, the number of output channels of ResNet, and the size of the BiLSTM hidden state are 20, 512, and 256, respectively. The setting of optimizer used in our experiments is also the same in [2,22] for fair comparisons.…”

Section: Implementation Detailsmentioning

confidence: 99%

“…(UDA) to couple the labeled source domain and unlabeled target domain to learn a domain-invariant model [11,12,13,14,15,16,18,22]. Our work aims to apply UDA with labeled synthetic text images and unlabeled real-world text images to learn an improved text recognition model in the target domain (i.e., real-world text images) by comparing the model learned from synthetic text images.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Sequence-To-Sequence Domain Adaptation Network for Robust Text Image Recognition

Zhang

Nie

Liu

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

129

View full text Add to dashboard Cite

Training recognition models with synthetic images have achieved remarkable results in text recognition. However, recognizing text from real-world images still faces challenges due to the domain shift between synthetic and real-world text images. One of the strategies to eliminate the domain difference without manual annotation is unsupervised domain adaptation (UDA). Due to the characteristic of sequential labeling tasks, most popular UDA methods cannot be directly applied to text recognition. To tackle this problem, we proposed a UDA method with minimizing latent entropy on sequence-to-sequence attention-based models with classbalanced self-paced learning. Our experiments show that our proposed framework achieves better recognition results than the existing methods on most UDA text recognition benchmarks. All codes are publicly available 1 .

show abstract

Multi-teacher Knowledge Distillation for End-to-End Text Image Machine Translation

Ma,

Zhang,

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Robust Text Image Recognition via Adversarial Sequence-to-Sequence Domain Adaptation

Cited by 23 publications

References 33 publications

[Retracted] Research on Recognition Effect of DSCN Network Structure in Hand‐Drawn Sketch

[Retracted] Research on Recognition Effect of DSCN Network Structure in Hand‐Drawn Sketch

Sequence-To-Sequence Domain Adaptation Network for Robust Text Image Recognition

Multi-teacher Knowledge Distillation for End-to-End Text Image Machine Translation

Contact Info

Product

Resources

About