SwinTextSpotter: Scene Text Spotting via Better Synergy between Text Detection and Text Recognition

Huang, Mingxin; Liu, Yuliang; Peng, Zhonghua; Liu, Chongyu; Lin, Dahua; Zhu, Shenggao; Yuan, Nicholas Jing; Ding, Kai; Jin, Lianwen

doi:10.1109/cvpr52688.2022.00455

Cited by 85 publications

(33 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Zhong et al [9] added a balance module to the decoder, which can reduce the impact of noise on characters. Huang et al [10] used the results of text detection to correct the loss of text recognition and achieved better recognition performance. He et al [11] employed the segmentation network to segment the characters, generated the corresponding graph node network according to the segmentation graph, and finally utilized the graph neural network for recognition.…”

Section: Deep Learning-based Text Recognition Methodsmentioning

confidence: 99%

“…where is the sigmoid function, and f 7×7 indicates that the convolutional kernel is a 7 × 7 convolution operation. For convenience, we set the parameters according to references [6] and [10]. Extensive experiments have proved that the parameter settings in references [6] and [10] can obtain satisfactory results in different computer vision tasks.…”

Section: Convolutional Block Attention Modulementioning

confidence: 99%

“…For convenience, we set the parameters according to references [6] and [10]. Extensive experiments have proved that the parameter settings in references [6] and [10] can obtain satisfactory results in different computer vision tasks. On this basis, we add channel attention and spatial attention to the network parameter settings.…”

Section: Convolutional Block Attention Modulementioning

confidence: 99%

“…Unlike traditional text recognition methods, deep learning methods [5][6][7][8][9][10][11] perform a top-down approach to recognize words or text lines directly, showing more promising results. Recent approaches [4,12] treat the STR problem as a sequence prediction problem, resulting in increased precision while eliminating the need for character-level annotation.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

NRSTRNet: A Novel Network for Noise-Robust Scene Text Recognition

Yue

Huang

Vong³

et al. 2023

Int J Comput Intell Syst

View full text Add to dashboard Cite

Scene text recognition (STR) has been widely applied in industrial and commercial fields. However, existing methods still face challenges when processing text images with defects such as low contrast, blur, low resolution, and insufficient illumination. These defects are common in actual situations because of diverse text backgrounds in natural scenes and limitations in shooting conditions. To address these challenges, we propose a novel network for noise-robust scene text recognition (NRSTRNet), which comprehensively suppresses the noise in the three critical steps of STR. Specifically, in the text feature extraction stage, NRSTRNet enhances the text-related features through the channel and spatial dimensions and disregards some disturbances from the non-text area, reducing the noise and redundancy in the input image. In the context encoding stage, fine-grained feature coding is proposed to effectively reduce the influence of previous noisy temporal features on current temporal features while simultaneously reducing the impact of partial noise on the overall encoding by sharing contextual feature encoding parameters. In the decoding stage, a self-attention module is added to enhance the connections between different temporal features, thereby leveraging the global information to obtain noise-resistant features. Through these approaches, NRSTRNet can enhance the local semantic information while considering the global semantic information. Experimental results show that the proposed NRSTRNet can improve the ability to characterize text images, enhance stability under the influence of noise, and achieve superior accuracy in text recognition. As a result, our model outperforms SOTA STR models on irregular text recognition benchmarks by 2% on average, and it is exceptionally robust when applied to noisy images.

show abstract

Section: Deep Learning-based Text Recognition Methodsmentioning

confidence: 99%

Section: Convolutional Block Attention Modulementioning

confidence: 99%

Section: Convolutional Block Attention Modulementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

NRSTRNet: A Novel Network for Noise-Robust Scene Text Recognition

Yue

Huang

Vong³

et al. 2023

Int J Comput Intell Syst

View full text Add to dashboard Cite

show abstract

“…al., 2022) They proposed SwinTextSpotter for text detection and text recognition. The overall architecture of SwinTextSpotter consists of four components: a backbone based on Swin-Transformer; a query-based text detector; a Recognition Conversion module to bridge the text detector and recognizer; and an attention-based recognizer [10]. This method gives a small percentage (66.9%) in determining the texts affiliated with the (Total Text) data set.…”

Section: Related Workmentioning

confidence: 99%

Scene Text detection and Recognition by Using Multi-Level Features Extractions Based on You Only Once Version Five (YOLOv5) and Maximally Stable Extremal Regions (MSERs) with Optical Character Recognition (OCR)

Almousawee,

El Abbadi

2022

AJEST

View full text Add to dashboard Cite

Textual information within scene images is very important in computer vision applications such as image retrieval based on its content, Tourist translator, and Navigation systems assistant. This paper presented a scene text recognition system based on YOLO. The main steps of the suggested methodology are: text detection and localization (using YOLOv5), text segmentation (using Morphological processing as a new method), features extraction (using MSERs), character segmentation and word segmentation (using bounding boxes with graph), and finally character recognition (using OCR). In this work we create a new dataset model that includes most of the text challenges such as Font type, Font size, Font color, and Font. The proposed system gives higher performance for detection, localization, and recognition when using dataset containing many challenges, the results were 80%, 96%, and 87.6 for precision, recall, and F-score respectively. Comparing with other similar works it was better. The accuracy of (OCR) is more than 99%.

show abstract