A robust arbitrary text detection system for natural scene images

By Anhar Risnumawan, Palaiahnakote Shivakumara, Chee Seng Chan, Chew Lim Tan. 2014

551 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Natural Language Processing

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

In Media Technology

In Natural Language Processing

Citations in the corpus

GIT: A Generative Image-to-text Transformer for Vision and Language (2022, methodology): “, IC13 [52], SVT [107], IIIT [80], IC15 [51], SVTP [86], CUTE [91] from the top to the bottom row, respectively.”
A Multi-Object Rectified Attention Network for Scene Text Recognition (2019, background): “CUTE80 [37] contains 80 high-resolution images taken in natural scenes.”
What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis (2019, background): “• CUTE80 (CT) [22] is collected from natural scenes and contains 288 cropped images for evaluation.”
Arbitrary-Oriented Scene Text Detection via Rotation Proposals (2017, background): “For example, [36] uses mutual magnitude symmetry and gradient vector symmetry to identify text pixel candidates regardless of the orientation, including curves from natural”
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding (2023, background): “• CT80 [57] contains 80 images for curved text OCR evaluation.”
Structured Multimodal Attentions for TextVQA (2020, methodology): “The robust transformer based network is trained on following datasets: IIIT 5K-Words [29], Street View Text [43], ICDAR 2013 [22], ICDAR 2015 [21], Street View Text Perspective [33], CUTE80 [36] and ArT [10].”
LVLM-EHub: A Comprehensive Evaluation Benchmark for Large Vision-Language Models (2023, unknown): “We evaluate this capability through Optical Characters Recognition (OCR) using twelve benchmarks, including IIIT5K [66], IC13 [67], IC15 [68], Total-Text [69], CUTE80 [70], SVT [71], SVTP [72], COCO-Text [73], Wor-dArt [74], CTW [75], HOST [76], and WOST [76].”
ImageBind-LLM: Multi-modality Instruction Tuning (2023, methodology): “We evaluate ImageBind-LLM on 12 representative OCR datasets, including IIIT5K [61], ICDAR 2013(IC13) [62], ICDAR 2015 (IC15) [63], Total-Text [64], CUTE80 [65], Street View Text (SVT) [66], SVTP-Perspective (SVTP) [67], COCO-Text [68], WordArt [69], SCUT-CTW1500 (CTW) [70], Heavily Occluded Scene Te”
MaskOCR: Text Recognition with Masked Encoder-Decoder Pretraining (2022, methodology): “We evaluate our model on six public scene text datasets: ICDAR 2013 (IC13) [26], Street View Text (SVT) [52], IIIT5K-Words (IIIT5K) [38]), ICDAR 2015 (IC15) [25], Street View Text-Perspective (SVTP) [39], and CUTE80 (CUTE) [43]).”
Towards Accurate Scene Text Recognition With Semantic Reasoning Networks (2020, background): “CUTE80 (CUTE) is proposed in [28] for curved text recognition.”
ASTER: An Attentional Scene Text Recognizer with Flexible Rectification (2019, unknown): “CUTE80 (CUTE) is proposed in [51].”
Show, Attend and Read: A Simple and Strong Baseline for Irregular Text Recognition (2018, background): “On irregular text datasets ( i.e ., IC15, SVTP, CT80 and COCO-T), our approach outperforms the compared methods by a large margin.”
ESIR: End-To-End Scene Text Recognition via Iterative Image Rectification (2018, background): “This can be observed more clearly over the two highly distorted datasets SVTP and CUTE as shown in Table 3.”
OCRBench: on the hidden mystery of OCR in large multimodal models (2023, unknown): “…datasets, including (1) Regular Text Recognition: IIIT5K [37], SVT [38], IC13 [39]; (2)Irregular Text Recognition: IC15 [40], SVTP [41], CT80 [42], COCOText(COCO) [43], SCUT-CTW1500 (CTW) [44], Total-Text (TT) [45]; (3) Occlusion Scene Text [46], encompassing weakly occluded scene text…”
Decoupled Attention Network for Text Recognition (2019, background): “CUTE80 focused on curved text, and consisted of 80 high-resolution images taken in natural scenes.”
Scene Text Detection and Recognition: The Deep Learning Era (2018, background): “ecognition. SVT28 The Street View Text (SVT) dataset [156], [157] is a collection of street view images. SVT has 350 images. It only has word-level annotations. CUTE80 (CUTE) 29 CUTE is proposed in [120]. The dataset focuses on curved text. It contains 80 high-resolution images taken in natural scen”
Symmetry-Constrained Rectification Network for Scene Text Recognition (2019, background): “%) and CUTE80 (+3.1%).”