检索结果-内蒙古大学图书馆

scene text segmentation via Multi-Task Cascade Transformer With Paired Data Synthesis

IEEE ACCESS 2023年 11卷 67791-67805页

作者： Dang, Quang-Vinh Lee, Guee-Sang Chonnam Natl Univ Dept Artificial Intelligence Convergence Gwangju 61186 South Korea

The scene text segmentation task provides a wide range of practical applications. However, the number of images in the available datasets for scene text segmentation is not large enough to effectively train deep learning-based models, leading to limited performance. To solve this problem, we employ paired data generation to secure sufficient data samples for text segmentation via text Image-conditional GANs. Furthermore, existing models implicitly model text attributes such as size, layout, font, and structure, which hinders their performance. To remedy this, we propose a Multi-task Cascade Transformer network that explicitly learns these attributes using large volumes of generated synthetic data. The transformer-based network includes two auxiliary tasks and one main task for text segmentation. The auxiliary tasks help the network learn text regions to focus on, as well as the structure of the text through different words and fonts, to support the main task. To bridge the gap between different datasets, we train the proposed network on paired synthetic data before fine-tuning it on real data. Our experiments on publicly available scene text segmentation datasets show that our method outperforms existing methods.

关键词： scene text segmentation paired data synthesis GANs transformer multi-task cascade

来源：评论

学校读者我要写书评

暂无评论

scene text segmentation BY PAIRED DATA SYNTHESIS 30

SCENE TEXT SEGMENTATION BY PAIRED DATA SYNTHESIS

引用

30th IEEE International Conference on Image Processing (ICIP)

作者： Quang-Vinh Dang Lee, Guee-Sang Chonnam Natl Univ Dept Artificial Intelligence Convergence Gwangju South Korea

ISBN: (纸本)9781728198354

scene text segmentation task has numerous practical applications. However, the number of images in the available datasets for scene text segmentation is too small to effectively train deep learning-based models, leading to limited performance. To solve this problem, we perform the segmentation in two aspects: paired data synthesis and methodology. The former is executed via the proposed text Image-conditional GANs to generate realistic paired data. We exploit real-world images by self-supervised pre-training scheme via inpainting approach before training the proposed GANs to produce realistic synthetic data. The latter is carried out by the proposed scene text segmentation network to optimize learning the generated paired data, called Multi-task Cascade Transformer. It includes two auxiliary tasks and one main task for text segmentation. The functions of the two auxiliary tasks are to learn the text region to focus on, together with learning the structure of text through their fonts, and then they support the main task. We implement three publicly available datasets for scene text segmentation: ICDAR13 FST, Total text, and textSeg datasets to demonstrate the effectiveness of our method. Our experimental result outperforms existing methods.

关键词： scene text segmentation Paired Data Synthesis GANs Transformer Multi-task Cascade

来源：评论

学校读者我要写书评

暂无评论

scene text segmentation with text-Focused Transformers 23

Scene Text Segmentation with Text-Focused Transformers

引用

31st ACM International Conference on Multimedia (MM)

作者： Yu, Haiyang Wang, Xiaocong Niu, Ke Li, Bin Xue, Xiangyang Fudan Univ Sch Comp Sci Shanghai Key Lab IIP Shanghai Peoples R China

ISBN: (纸本)9798400701085

text segmentation is a crucial aspect of various text-related tasks, including text erasing, text editing, and font style transfer. In recent years, multiple text segmentation datasets, such as textSeg focusing on Latin text segmentation and BTS on bilingual text segmentation, have been proposed. However, existing methods either disregard the annotations of text location or directly use pre-trained text detectors. In general, these methods cannot fully utilize the annotations of text location in the datasets. To explicitly incorporate text location information to guide text segmentation, we propose an end-to-end text-focused segmentation framework, where text detection and segmentation are jointly optimized. In the proposed framework, we first extract multi-level global visual features through residual convolution blocks and then predict the mask of text areas using a text detection head. Subsequently, we develop a text-focused module that compels the model to pay more attention to text areas. Specifically, we introduce two types of attention masks to extract corresponding features: text-aware and instance-aware features. Finally, we employ hierarchical Transformer encoders to fuse multi-level features and predict the text mask with a text segmentation head. To evaluate the effectiveness of our method, we conduct experiments on six text segmentation benchmarks. The experimental results demonstrate that the proposed method outperforms the previous state-of-the-art (SOTA) methods by a clear margin in most cases. The code and supplementary materials are available at https://***/FudanVI/FudanOCR/tree/main/text-focused-Transformers.

关键词： scene text segmentation text-Focused Transformer End-to-End Framework

来源：评论

学校读者我要写书评

暂无评论

Looking from a Higher-Level Perspective: Attention and Recognition Enhanced Multi-scale scene text segmentation 16th

Looking from a Higher-Level Perspective: Attention and Recog...

引用

16th Asian Conference on Computer Vision (ACCV)

作者： Ren, Yujin Zhang, Jiaxin Chen, Bangdong Zhang, Xiaoyi Jin, Lianwen South China Univ Technol Guangzhou Peoples R China SCUT Zhuhai Inst Modern Ind Innovat Zhuhai Peoples R China

ISBN: (纸本)9783031262920;9783031262937

scene text segmentation, which aims to generate pixel-level text masks, is an integral part of many fine-grained text tasks, such as text editing and text removal. Multi-scale irregular scene texts are often trapped in complex background noise around the image, and their textures are diverse and sometimes even similar to those of the background. These specific problems bring challenges that make general segmentation methods ineffective in the context of scene text. To tackle the aforementioned issues, we propose a new scene text segmentation pipeline called Attention and Recognition enhanced Multi-scale segmentation Network (ARM-Net), which consists of three main components: text segmentation Module (TSM) generates rectangular receptive fields of various sizes to fit scene text and integrate global information adequately;Dual Perceptual Decoder (DPD) strengthens the connection between pixels that belong to the same category from the spatial and channel perspective simultaneously during upsampling, and Recognition Enhanced Module (REM) provides text attention maps as a prior for the segmentation network, which can inherently distinguish text from background noise. Via extensive experiments, we demonstrate the effectiveness of each module of ARM-Net, and its performance surpasses that of existing state-of-the-art scene text segmentation methods. We also show that the pixel-level mask produced by our method can further improve the performance of text removal and scene text recognition.

关键词： scene text segmentation Deep neural network

来源：评论

学校读者我要写书评

暂无评论

Efficient scene text segmentation via Gate Mechanism and Multi-Scale Feature Fusion 3

Efficient Scene Text Segmentation via Gate Mechanism and Mul...

引用

3rd International Conference on Electronic Information Engineering and Computer Communication, EIECC 2023

作者： Zhang, Guodong Lai, Qinling State Grid Fujian Electric Power Co. Ltd. Changle District Power Supply Company Fuzhou City China Fuzhou Power Supply Company State Grid Fujian Electric Power Co. Lt Fuzhou China

ISBN: (纸本)9798350359961

In addressing the challenges posed by current text segmentation methods when dealing with multi-scale and irregular scene text, this paper presents an innovative approach to scene text segmentation. The proposed method leverages multiple Gated Fully Fusion Modules, facilitating adaptive weight adjustments for features at various levels and selective information aggregation. This ensures enhanced capture of multiscale text information. Additionally, a Multi-Scale Feature Fusion Module further improves information capture across different scales. The incorporation of a text Prior Module and an Enhancement Module focuses on learning text prior information and edge features, respectively, refining the precise recognition and segmentation capabilities of multi-scale text targets. Experimental evaluations on the textSegt dataset and the Bilingual text segmentation dataset validate the effectiveness of the proposed algorithm. Results demonstrate outstanding performance in handling text of various languages and sizes, highlighting robust adaptability and generalization capabilities. This work presents a compelling solution for scene text segmentation tasks. © 2023 IEEE.

关键词： Attention mechanism Deep learning Gate mechanism Multi-scale feature fusion scene text segmentation

来源：评论

学校读者我要写书评

暂无评论

Enhancing scene text segmentation through Subtask Decomposition 2

Enhancing Scene Text Segmentation through Subtask Decomposit...

引用

2nd International Conference on Image Processing, Computer Vision and Machine Learning, ICICML 2023

作者： Wang, Yong Chen, Youguang East China Normal University School of Data Science and Engineering Shanghai China

ISBN: (纸本)9798350331417

The field of image processing widely utilizes scene text segmentation technology, with applications extending to image editing and font style transfer. These applications enhance image understanding quality and aid in boosting the performance of numerous computer vision tasks. The advent and progression of deep learning have led to substantial advancements in scene text segmentation technology. However, the limited size of existing scene text segmentation datasets constrains the performance of models. Therefore, we propose an algorithm for synthetic segmentation data. We first pretrain the model using large-scale synthetic data, then fine-tune it on the target dataset to address the issue of limited dataset size. Existing models employ end-to-end segmentation, which presents challenges in segmentation. We propose a scene text segmentation method. By decomposing the segmentation task into subtasks and solving them one by one, the complexity of the task can be reduced compared to direct segmentation of the entire image significantly improving the segmentation effect. The proposed method consists of three modules: a fragment crop module, a fragment segmentation module, and a fragment combination module. The fragment crop module is composed of an additional corp layer added after DBnet. The fragment segmentation module can be embedded with various segmentation methods. The fragment combination module uses the maximum pixel value pasting algorithm to combine the segmented fragments. We call this method Crop-segmentation-Combination Framework (CSCF). We conducted experiments on the ICDAR 2013 and textSeg datasets. The CSCF, embedded in Unet within the segment segmentation module, enhanced the text segmentation IoU by 5.80% on the ICDAR 2013 test dataset. Our suggested approach has been shown to notably enhance the efficiency of scene text segmentation. © 2023 IEEE.

关键词： Image segmentation scene text segmentation Synthetic data synthetic image generations text segmentation

来源：评论

学校读者我要写书评

暂无评论

Self-supervised scene text segmentation with Object-centric Layered Representations Augmented by text Regions 22

Self-supervised Scene Text Segmentation with Object-centric ...

引用

30th ACM International Conference on Multimedia (MM)

作者： Wang, Yibo Ye, Yunhu Mao, Yuanpeng Yu, Yanwei Song, Yuanping Univ Sci & Technol China Sch Software Engn Hefei Peoples R China Univ Sci & Technol China Suzhou Inst Adv Res Hefei Peoples R China

ISBN: (纸本)9781450392037

text segmentation tasks have a very wide range of application values, such as image editing, style transfer, watermark removal, etc. However, existing public datasets are of poor quality of pixel-level labels that have been shown to be notoriously costly to acquire, both in terms of money and time. At the same time, when pre-training is performed on synthetic datasets, the data distribution of the synthetic datasets is far from the data distribution in the real scene. These all pose a huge challenge to the current pixel-level text segmentation algorithms. To alleviate the above problems, we propose a self-supervised scene text segmentation algorithm with layered decoupling of representations derived from the object-centric manner to segment images into texts and background. In our method, we propose two novel designs which include Region Query Module and Representation Consistency Constraints adapting to the unique properties of text as complements to Auto Encoder, which improves the network's sensitivity to texts. For this unique design, we treat the polygon-level masks predicted by the text localization model as extra input information, and neither utilize any pixel-level mask annotations for training stage nor pretrain on synthetic datasets. Extensive experiments show the effectiveness of the method proposed. On several public scene text datasets, our method outperforms the state-of-the-art unsupervised segmentation algorithms.

关键词： scene text segmentation self-supervised learning layered representation object-centric

来源：评论

学校读者我要写书评

暂无评论

Leveraging text Signed Distance Function Map for Boundary-Aware Guidance in scene text segmentation

引用

IEEE ACCESS 2025年 13卷 78850-78863页

作者： Kim, Ho Jun Kim, Hak Gu Chung Ang Univ Dept Image Sci & Arts GSAIM Seoul 06974 South Korea Chung Ang Univ Dept Metaverse Convergence GSAIM Seoul 06974 South Korea

scene text segmentation is to predict pixel-wise text regions from an image, enabling in-image text editing or removal. One of the primary challenges is to remove noises including non-text regions and predict intricate text boundaries. To deal with that, traditional approaches utilize a text detection or recognition module explicitly. However, they are likely to highlight noise around the text. Because they did not sufficiently consider the boundaries of text, they fail to accurately predict the fine details of text. In this paper, we introduce leveraging text signed distance function (SDF) map, which encodes distance information from text boundaries, in scene text segmentation to explicitly provide text boundary information. By spatial cross attention mechanism, we encode the text-attended feature from the text SDF map. Then, both visual and text-attended features are utilized to decode the text segmentation map. Our approach not only mitigates confusion between text and complex backgrounds by eliminating false positives such as logos and texture blobs located far from the text, but also effectively captures fine details of complex text patterns by leveraging text boundary information. Extensive experiments demonstrate that leveraging text SDF map in scene text segmentation provides superior performances on various scene text segmentation datasets.

关键词： Image segmentation Visualization Feature extraction Decoding Semantics Transformers text recognition Noise Instance segmentation Accuracy scene text segmentation signed distance function optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Pixel-Level scene text segmentation by Mutually Guided Network

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2021年 30卷 8212-8221页

作者： Wang, Chuan Zhao, Shan Zhu, Li Luo, Kunming Guo, Yanwen Wang, Jue Liu, Shuaicheng Megvii Technol Beijing 100190 Peoples R China Chongqing Univ Posts & Telecommun Sch Commun & Informat Engn Chongqing 400065 Peoples R China Nanjing Univ Natl Key Lab Novel Software Technol Nanjing 210023 Peoples R China Univ Elect Sci & Technol China Sch Informat & Commun Engn Chengdu 611731 Peoples R China

In this paper we present a new data-driven method for pixel-level scene text segmentation from a single natural image. Although scene text detection, i.e. producing a text region mask, has been well studied in the past decade, pixel-level text segmentation is still an open problem due to the lack of massive pixel-level labeled data for supervised training. To tackle this issue, we incorporate text region mask as an auxiliary data into this task, considering acquiring large-scale of labeled text region mask is commonly less expensive and time-consuming. To be specific, we propose a mutually guided network which produces a polygon-level mask in one branch and a pixel-level text mask in the other. The two branches' outputs serve as guidance for each other and the whole network is trained via a semi-supervised learning strategy. Extensive experiments are conducted to demonstrate the effectiveness of our mutually guided network, and experimental results show our network outperforms the state-of-the-art in pixel-level scene text segmentation. We also demonstrate the mask produced by our network could improve the text recognition performance besides the trivial image editing application.

关键词： Image segmentation Task analysis Training Decoding Feature extraction text recognition Convolutional neural networks Semi-supervised scene text segmentation mutually guided network

来源：评论

学校读者我要写书评

暂无评论

COCO_TS Dataset: Pixel-Level Annotations Based on Weak Supervision for scene text segmentation 28th

COCO_TS Dataset: Pixel-Level Annotations Based on Weak Super...

引用

28th International Conference on Artificial Neural Networks (ICANN)

作者： Bonechi, Simone Andreini, Paolo Bianchini, Monica Scarselli, Franco Univ Siena DIISM Via Roma 56 Siena Italy

ISBN: (纸本)9783030305086;9783030305079

The absence of large scale datasets with pixel-level supervisions is a significant obstacle for the training of deep convolutional networks for scene text segmentation. For this reason, synthetic data generation is normally employed to enlarge the training dataset. Nonetheless, synthetic data cannot reproduce the complexity and variability of natural images. In this paper, a weakly supervised learning approach is used to reduce the shift between training on real and synthetic data. Pixel-level supervisions for a text detection dataset (i.e. where only bounding-box annotations are available) are generated. In particular, the COCO-text-segmentation (COCO_TS) dataset, which provides pixel-level supervisions for the COCO-text dataset, is created and released. The generated annotations are used to train a deep convolutional neural network for semantic segmentation. Experiments show that the proposed dataset can be used instead of synthetic data, allowing us to use only a fraction of the training samples and significantly improving the performances.

关键词： scene text segmentation Weakly supervised learning Bounding-box supervision Convolutional Neural Networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：