A Unified Framework for Multioriented Text Detection and Recognition

By Cong Yao, Xiang Bai. IEEE Transactions on Image Processing, 2014

300 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/2950be66e7b4c94dbb16e3319d8bece5da4e799f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Xiang Bai
2. Chunhua Shen
3. Luc Van Gool
4. Dacheng Tao
5. Cong Yao
6. Yu Qiao
7. Xiaogang Wang
8. Alan L. Yuille

Citations in the corpus

• Integrating Scene Text and Visual Appearance for Fine-Grained Image Classification (2017, methodology): “methods [39], [40], [56]–[58] and word-based methods [11], [45], [59] are the two mainstream recognition approaches.”
• Arbitrary-Oriented Scene Text Detection via Rotation Proposals (2017, methodology): “We train a baseline experiment on the ICDAR2015 benchmark using the same strategy used for MSRA-TD500.”
• Scene Text Detection and Recognition: The Deep Learning Era (2018, methodology): “HUST-TR400 [165] are collected in the same way as MSRATD500, usually used as a supplement to the MSRA-TD500 training data.”
• A Fast Uyghur Text Detector for Complex Background Images (2018, background): “T EXT localization in images with complex background is crucial to many real-world vision tasks, such as optical character recognition (OCR), text image analysis and scene understanding [1].”
• Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes (2018, methodology): “Inspired by [51], we propose a weighted edit distance algorithm.”
• Symmetry-based text line detection in natural scenes (2015, unknown): “[39] 0.”
• An End-to-End Trainable Neural Network for Image-Based Sequence Recognition and Its Application to Scene Text Recognition (2015, background): “[16] unified text detection and recognition by sharing features between the two tasks.”
• ASTER: An Attentional Scene Text Recognizer with Flexible Rectification (2019, methodology): “Representative proposal extraction methods include Maximally Stable Extremal Regions (MSER) [43], [46], Extremal Regions (ER) [47], and Stroke Width Transform (SWT) [14], [65], [66].”
• Scene text detection and recognition: recent advances and future trends (2015, methodology): “SWT [9] and MSER [10] are two representative methods in the field of scene text detection, which constitute the basis of a lot of subsequent works [12–14, 29, 30, 34, 48, 49].”
• Efficient and Accurate Arbitrary-Shaped Text Detection With Pixel Aggregation Network (2019, methodology): “Because the training set is rather small, we follow the previous works [57, 37, 35] to include the 400 images from HUST-TR400 [50] as training data.”
• Real-time Scene Text Detection with Differentiable Binarization (2019, methodology): “Following the previous methods (Zhou et al. 2017; Lyu et al. 2018b; Long et al. 2018), we include extra 400 training images from HUSTTR400 (Yao, Bai, and Liu 2014).”
• Text/non-text image classification in the wild with convolutional neural networks (2017, background): “To precisely localize text in an image, algorithms like [4, 5, 6, 7, 8] typically require searching a large set of text-line or character candidates, or dense image patches.”
• EAST: An Efficient and Accurate Scene Text Detector (2017, unknown): “Since the number of training images is too few to learn a deep model, we also harness 400 images from HUSTTR400 dataset [39] as training data.”
• TextBoxes++: A Single-Shot Oriented Scene Text Detector (2018, background): “A unified framework for multi-oriented text detection and recognition is proposed in [37].”
• Gliding Vertex on the Horizontal Bounding Box for Multi-Oriented Object Detection (2019, methodology): “Since the training set is rather small, following other methods, we also use HUST-TR400 [38] during training.”
• Real-Time Scene Text Detection With Differentiable Binarization and Adaptive Scale Fusion (2022, methodology): “Following the previous methods [37], [39], [65], we include extra 400 training images from HUSTTR400 [58].”