Home › Datasets › task › Scene Text Detection
Scene Text Detection datasets
archive 2025-07-28
17 datasets carry the task tag "Scene Text Detection" (the task itself: Scene Text Detection), ordered by the archive's paper count. Page 1 of 1: 17 shown of 17. Facet routes are this site's own (the archive records the tag string, not a page).
The archive holds 12,214 dataset rows; 12,172 are listed. 6 are withheld from every listing and count here as vandalised before snapshot (6 with contact-centre spam in the title, 0 with a spam description on a row that has no homepage, no paper and no papers counted; none with more than 1 paper, 0 with a benchmark), listed in withheld.json; 1 listed row carries a vandalised description, withheld on its page. This gate never withholds a row with a homepage or a paper that resolves, and a clean description; the content rules below withhold a row whose name is spam whatever else it carries. The gate is a phrase list: these are the rows it caught, not a claim that the rest is clean. Before that gate, the site's content rules withhold 36 more rows (invite-code, gambling, travel-booking, contact-centre and similar spam in the name or on a row with nothing real behind it); they have no page and are listed in withheld.json.
Filter 51 task tags shown of 3,717, by dataset count; the full filter by modality, task and language is on /datasets
Scene Text Detection datasets 1–17 of 17
The ICDAR 2013 dataset consists of 229 training images and 233 testing images, with word-level annotations provided.
246 papers · 3 benchmarks
Total-Text is a text detection dataset that consists of 1,555 images with a variety of text types including horizontal, multi-oriented, and curved text instances.
156 papers · 2 benchmarks
The MSRA-TD500 dataset is a text detection dataset that contains 300 training images and 200 test images.
124 papers · 1 benchmark
The COCO-Text dataset is a dataset for text detection and recognition.
89 papers · 2 benchmarks
ICDAR 2015 was a scene text detection used for the ICDAR 2015 conference.
52 papers · 2 benchmarks
The SCUT-CTW1500 dataset contains 1,500 images: 1,000 for training and 500 for testing.
44 papers · 3 benchmarks
TextOCR is a dataset to benchmark text recognition on arbitrary shaped scene-text.
37 papers · 0 benchmarks
RCTW-17 (Reading Chinese Text in the Wild)
Features a large-scale dataset with 12,263 annotated images.
20 papers · 0 benchmarks
ICDAR2017 is a dataset for scene text detection.
18 papers · 1 benchmark
Chinese Text in the Wild is a dataset of Chinese text with about 1 million Chinese characters from 3850 unique ones annotated by experts in over 30000 street view images.
5 papers · 0 benchmarks
The PKU dataset has almost 4,000 images categorized into five groups (G1-G5) that show different situations.
2 papers · 0 benchmarks
A newly developed natural scene text dataset of Chinese shop signs in street views.
2 papers · 0 benchmarks
The UrduDoc Dataset is a benchmark dataset for Urdu text line detection in scanned documents.
1 paper · 1 benchmark
CNTD (Chinese and Naxi text detection)
Chinese and Naxi scene text detection data set, labelme to json.
0 papers · 0 benchmarks
This dataset is an extremely challenging set of over 20,000+ original Number plate images captured and crowdsourced from over 700+ urban and rural areas, where each image is manually reviewed and verified by computer vision professionals…
0 papers · 0 benchmarks
Media-Text (MediaText: a media industry-based dataset for scene text detetcion)
Media-Text dataset comprising images of banners, posters, covers and another images characterised for media industry.
0 papers · 0 benchmarks
This dataset contains images and annotations for scene text detection and recognition.
0 papers · 0 benchmarks
Paper counts and descriptions are the archive's, frozen 2025-07-28; no citation counts, no stars, no trending. Sorting by "most cited" or "newest" was a live-site feature the archive does not carry.