Total-Text: A Comprehensive Dataset for Scene Text Detection and Recognition

Text in curve orientation, despite being one of the common text orientations in real world environment, has close to zero existence in well received scene text datasets such as ICDAR'13 and MSRA-TD500. The main motivation of Total-Text is to fill this gap and facilitate a new research direction for the scene text community. On top of conventional horizontal and multi-oriented text, it features curved-oriented text. Total-Text is highly diversified in orientations, more than half of its images have a combination of more than two orientations. Recently, a new breed of solutions that casted text detection as a segmentation problem has demonstrated their effectiveness against multi-oriented text. In order to evaluate its robustness against curved text, we fine-tuned DeconvNet and benchmark it on Total-Text. Total-Text with its annotation is available at https://github.com/cs-chan/Total-Text-Dataset.

• Publication year

  2017

• Venue

  IEEE International Conference on Document Analysis and Recognition

• Publication date

  2017-10-28

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/ICDAR.2017.157arXiv 1710.10400
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• Fine-tuned DeconvNet was benchmarked on Total-Text to evaluate segmentation-based text detection robustness against curved text.

  Confidence 0.88

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• More than half of Total-Text images contain a combination of more than two text orientations, making it highly diversified.

  Confidence 0.97

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• Total-Text fills the gap in existing scene text datasets by providing curved-oriented text annotations alongside conventional horizontal and multi-oriented text.

  Confidence 0.93

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review

• curved-oriented text

  phenomenon, data characteristic

  Text arranged along curved paths, a common real-world text orientation underrepresented in prior scene text datasets.

  Aliases: curve text, curved text

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• deconvnet

  model, baseline

  A deconvolutional neural network fine-tuned in this work to perform segmentation-based scene text detection on Total-Text.

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• icdar'13

  dataset, baseline

  A widely used scene text benchmark dataset that lacks curved-text examples, noted here as a gap Total-Text addresses.

  Aliases: ICDAR 2013

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• msra-td500

  dataset, baseline

  An existing multi-oriented scene text dataset that, like ICDAR'13, does not include curved text.

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• scene text dataset

  dataset, evaluation setting

  A labeled image collection used for training and evaluating text detection and recognition systems in natural scenes.

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• segmentation-based text detection

  method

  An approach that formulates text detection as a pixel-level segmentation problem rather than bounding-box regression.

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review
• total-text

  dataset, resource

  A comprehensive scene text dataset introduced to include curved-oriented text in addition to horizontal and multi-oriented text.

  뀨 (7c402c1b98) extractionq (76h6bfydm6) reviewAll you need is Python (5d7gwfm5zu) review박진우 (dztg5apj7m) review

• Detecting Oriented Text in Natural Images by Linking Segments

  2017cited by this paper
• Robust Text Detection in Natural Scene Images

  2016cited by this paper
• Multi-oriented Text Detection with Fully Convolutional Networks

  2016influential reference
• COCO-Text: Dataset and Benchmark for Text Detection and Recognition in Natural Images

  2016cited by this paper
• Scene Text Detection via Holistic, Multi-Channel Prediction

  2016cited by this paper
• Accurate Text Localization in Natural Image with Cascaded Convolutional Text Network

  2016influential reference
• Multi-Orientation Scene Text Detection with Adaptive Clustering

  2015cited by this paper
• Learning Deconvolution Network for Semantic Segmentation

  2015influential reference
• ICDAR 2015 competition on Robust Reading

  2015influential reference
• Symmetry-based text line detection in natural scenes

  2015cited by this paper
• Robust Scene Text Detection with Convolution Neural Network Induced MSER Trees

  2014influential reference
• Fully convolutional networks for semantic segmentation

  2014influential reference
• A robust arbitrary text detection system for natural scene images

  2014influential reference
• Deep Features for Text Spotting

  2014cited by this paper
• Text Localization in Natural Images Using Stroke Feature Transform and Text Covariance Descriptors

  2013cited by this paper
• Scene Text Localization and Recognition with Oriented Stroke Detection

  2013cited by this paper
• ICDAR 2013 Robust Reading Competition

  2013cited by this paper
• End-to-end text recognition with convolutional neural networks

  2012cited by this paper
• Detecting texts of arbitrary orientations in natural images

  2012influential reference
• Detecting text in natural scenes with stroke width transform

  2010cited by this paper
• Object count/area graphs for the evaluation of object detection and segmentation algorithms

  2006cited by this paper
• Robust wide-baseline stereo from maximally stable extremal regions

  2004cited by this paper

• Text-Pass Filter: An Efficient Scene Text Detector

  2026cites this paper
• PositionOCR: Augmenting Positional Awareness in Multi-Modal Models via Hybrid Specialist Integration

  2026influential citation
• TiCLS : Tightly Coupled Language Text Spotter

  2026cites this paper
• A Self-Mutual Learning Framework Based on Knowledge Distillation for Scene Text Detection

  2026cites this paper
• ETLRS: An End-to-End Text Spotting Framework With Spatial and Directional Perception and Cross-Task Feature Synergy

  2026cites this paper
• Balancing Optimization Strategies and Practical Goals: An Efficient Scene Text Detector

  2026cites this paper
• Improving Pseudo-Labeling by Dynamic Confidence Calibration for Semi-Supervised Sequence Recognition

  2026cites this paper
• From Diffusion to Decision: A Diffusion-ReRanking in Scene Text Detection

  2025cites this paper
• Text images processing system using artificial intelligence models

  2025cites this paper
• Multilingual visual question answering for visually impaired people

  2025cites this paper
• A Token-level Text Image Foundation Model for Document Understanding

  2025cites this paper
• LRANet++: Low-Rank Approximation Network for Accurate and Efficient Text Spotting

  2025cites this paper
• A Comprehensive Survey of Transformers in Text Recognition: Techniques, Challenges, and Future Directions

  2025cites this paper
• Pull Pole Points to Text Contour by Magnetism: A Real-Time Scene Text Detector

  2025cites this paper
• Joint Low-level and High-level Textual Representation Learning with Multiple Masking Strategies

  2025cites this paper
• Bharat Scene Text: A Novel Comprehensive Dataset and Benchmark for Indian Language Scene Text Understanding

  2025cites this paper
• A Novel Framework for Robust Scene Text Localization: The OmniTextNet Paradigm

  2025cites this paper
• Bright to Dark: Stage-wise Bilevel Knowledge Transfer for Seeing Text in the Dark

  2025cites this paper
• Arbitrary-Shaped Text Detection with Hierarchical Feature Refinement and Nonlinear-Enhanced Transformers

  2025cites this paper
• Implicit Feature Fusion Function in Scene Text Detection

  2025cites this paper
• AI-Enabled Visual Assistance with Integrated Reading for People with Total Blindness: A Review

  2025influential citation
• A Text Detection Method Based on Multiscale Selective Fusion Feature Pyramid and Multisemantic Spatial Network for Visual IoT

  2025cites this paper
• A Large-scale Dataset for Robust Complex Anime Scene Text Detection

  2025cites this paper
• LICS: Locating Inter-Character Spaces for Multilingual Scene Text Detection

  2025cites this paper
• ASTD-ABC: Arbitrary Shape Text Detection with Adaptive Points and B-Spline Curves

  2025cites this paper
• TextMamba: Scene Text Detector with Mamba

  2025cites this paper
• Toward AI-Enabled Approach for Urdu Text Recognition: A Legacy for Urdu Image Apprehension

  2025cites this paper
• Automated document processing system for government agencies using DBNET++ and BART models

  2025cites this paper
• Leveraging Text Signed Distance Function Map for Boundary-Aware Guidance in Scene Text Segmentation

  2025influential citation
• A Comparative Analysis of OCR Models on Diverse Datasets: Insights from Memes and Hiertext Dataset

  2025cites this paper
• Image Retargeting based on Text Region Awareness

  2025cites this paper
• Compensating for the Incomplete With the Complete: An Efficient Scene Text Detector

  2025cites this paper
• ContraText-DETR: Boosting Industrial Scene Text Detection Based on Contrastive Learning and Synthetic Low-Contrast Text

  2025cites this paper
• A text clarification and deep relational reasoning method for Mongolian-Chinese bilingual arbitrary-shaped scene text detection

  2025cites this paper
• Starnet-tex: lightweight real-time text detection via middle-order enhanced contextual attention

  2025cites this paper
• Quantifying the Effects of Image Degradation on LVLM Benchmark Results Systematically

  2025cites this paper
• SEMFNet: Multi-path dynamic fusion text detection model combined with scale estimation module

  2025cites this paper
• AFFIR: Dual-Modal Attention Feature Fusion for Scene Text Image Retargeting

  2025cites this paper
• Advances in Text Detection and Recognition in Multi View Image Scenes

  2025cites this paper
• A Two-Stage End-to-End Framework for Robust Scene Text Spotting with Self-Calibrated Detection and Contextual Recognition

  2025cites this paper
• An efficient lightweight arbitrary scene text detection and segmentation system for low-end computing devices

  2025cites this paper
• A Comprehensive Review of Digital Watermarking techniques: Applications, Characteristics, Classification, and Related Aspects

  2025cites this paper
• ConText: Driving In-context Learning for Text Removal and Segmentation

  2025cites this paper
• GP-PSENet: a group-related dilated and a parallel extensional dilation-wise residual encoder for scene text detection

  2025cites this paper
• FSANet: Feature shuffle and adaptive channel attention network for arbitrary shape scene text detection

  2025cites this paper
• Scene Text Detection in Foggy Weather Utilizing Knowledge Distillation of Diffusion Models

  2025cites this paper
• Advancing Sequential Numerical Prediction in Autoregressive Models

  2025cites this paper
• A Bilingual, Open World Video Text Dataset and Real-Time Video Text Spotting With Contrastive Learning

  2025cites this paper
• The Devil is in Fine-tuning and Long-tailed Problems:A New Benchmark for Scene Text Detection

  2025cites this paper
• SynthBendText3D: a framework for generating scene text data in arbitrary orientations using a 3D graphics engine

  2025influential citation
• CCDPlus: Towards Accurate Character to Character Distillation for Text Recognition

  2025cites this paper
• Multimodal Large Language Models for Text-rich Image Understanding: A Comprehensive Review

  2025cites this paper
• Character Can Speak Directly: An End-to-End Character Region Excavation Network for Scene Text Spotting

  2025cites this paper
• S3INet: Semantic-Information Space Sharing Interaction Network for Arbitrary Shape Text Detection

  2025influential citation
• A Context-Driven Training-Free Network for Lightweight Scene Text Segmentation and Recognition

  2025influential citation
• Gender Classification From Pashto Handwritten Text Images

  2025cites this paper
• Marten: Visual Question Answering with Mask Generation for Multi-modal Document Understanding

  2025cites this paper
• ClapperText: A Benchmark for Text Recognition in Low-Resource Archival Documents

  2025cites this paper
• Edge Approximation Text Detector

  2025cites this paper
• Optimizing Text Recognition in Mechanical Drawings: A Comprehensive Approach

  2025cites this paper
• DocSAM: Unified Document Image Segmentation via Query Decomposition and Heterogeneous Mixed Learning

  2025cites this paper
• TSAL: Few-shot Text Segmentation Based on Attribute Learning

  2025cites this paper
• Perceive Anything: Recognize, Explain, Caption, and Segment Anything in Images and Videos

  2025cites this paper
• PIFText: Progressive Injection Fusion for Text Detection in Autonomous Driving Scenarios

  2025cites this paper
• Development of OCR Service for Page-Level Recognition for Camera-Captured Document Images

  2025cites this paper
• TextSR: Diffusion Super-Resolution with Multilingual OCR Guidance

  2025cites this paper
• Text-Aware Image Restoration with Diffusion Models

  2025cites this paper
• Text Extraction from Video Using Generative AI Models: A Multimodal Approach

  2025cites this paper
• Font-Agent: Enhancing Font Understanding with Large Language Models

  2025cites this paper
• PTDR: A Real-World and Synthetic Benchmark Dataset for Persian Scene and Document Text Detection and Recognition

  2025cites this paper
• Curved text revealed: enhancing detection with multidirectional pooling and spatial awareness

  2025cites this paper
• TiTDet: A tiny text detector with scale-sensitive loss and effective fusion factor

  2024cites this paper
• Scene Text Detection using Hyperbolic Tangent Binarization and CLIP

  2024cites this paper
• Real-time arbitrarily-shaped scene text spotting with weakly supervised points

  2024cites this paper
• Seeing Text in the Dark: Algorithm and Benchmark

  2024cites this paper
• Draw-and-Understand: Leveraging Visual Prompts to Enable MLLMs to Comprehend What You Want

  2024cites this paper
• Arbitrary-Shaped Text Detection via Ordered Construction of Convex Hull Vertices

  2024cites this paper
• TextBlockV2: Towards Precise-Detection-Free Scene Text Spotting with Pre-trained Language Model

  2024cites this paper
• YOLO-based Object Detection Models: A Review and its Applications

  2024cites this paper
• TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document

  2024cites this paper
• Beyond the Mud: Datasets and Benchmarks for Computer Vision in Off-Road Racing

  2024cites this paper
• Combining Swin Transformer and Attention-Weighted Fusion for Scene Text Detection

  2024cites this paper
• OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning

  2024cites this paper
• A real-time arbitrary-shape text detector

  2024cites this paper
• Efficiently Leveraging Linguistic Priors for Scene Text Spotting

  2024influential citation
• Hi-SAM: Marrying Segment Anything Model for Hierarchical Text Segmentation

  2024influential citation
• TTS: Hilbert Transform-Based Generative Adversarial Network for Tattoo and Scene Text Spotting

  2024cites this paper
• Evaluation of Open-Source OCR Libraries for Scene Text Recognition in the Presence of Fisheye Distortion

  2024influential citation
• Turkish scene text recognition: Introducing extensive real and synthetic datasets and a novel recognition model

  2024cites this paper
• Dynamic Relation Transformer for Contextual Text Block Detection

  2024cites this paper
• Partial Scene Text Retrieval

  2024cites this paper
• CC-OCR: A Comprehensive and Challenging OCR Benchmark for Evaluating Large Multimodal Models in Literacy

  2024cites this paper
• Buffer-text: Detecting arbitrary shaped text in natural scene image

  2024cites this paper
• Relational Contrastive Learning and Masked Image Modeling for Scene Text Recognition

  2024cites this paper
• Scene Text Detection Based on Bi-directional Feature Pyramid Network

  2024cites this paper
• Double supervision for scene text detection and recognition based on BMINet

  2024cites this paper
• Artistic-style text detector and a new Movie-Poster dataset

  2024cites this paper
• BPDO: Boundary Points Dynamic Optimization for Arbitrary Shape Scene Text Detection

  2024cites this paper
• When IC meets text: Towards a rich annotated integrated circuit text dataset

  2024cites this paper
• Document Parsing Unveiled: Techniques, Challenges, and Prospects for Structured Information Extraction

  2024cites this paper