OCR Datasets

This dataset contains 4,995 Vietnamese OCR images with annotations and text transcriptions. The data includes 258 natural scene images, 2,553 Internet images, and 2,184 document images. For line-level content annotation, quadrilateral bounding box annotations and text transcriptions are provided. For column-level content annotation, column-level quadrilateral bounding box annotation and text transcription are provided. The data can be used for tasks such as Vietnamese recognition in multiple scenes.

This dataset contains 3,506 Hindi OCR images, including 2,056 images of natural scenes, 1,103 Internet images and 347 document images. For line-level content annotation, line-level quadrilateral bounding box annotation and test transcription was adpoted; for column-level content annotation, column-level quadrilateral bounding box annotation and text transcription was adpoted. The dataset can be used for Hindi OCR, Hindi character recognition, and text detection tasks across multiple real-world scenes.

57,645 Images - Vertical OCR Data in Text Scenes. The collecting scenes of this dataset include street scenes, plaques, billboards, posters, decorations, art lettering, magazine covers etc. The language distribution includes Chinese and a few English. In this dataset, vertical -level rectangular bounding box (polygonal bounding box, parallelogram bounding box) annotation and transcription for the texts; non-vertical rectangular bounding box (polygonal bounding box, parallelogram bounding box) annotation and transcription for the texts. This dataset can be used for tasks such as multiple vertical text scenes OCR.

This dataset contains 14,980 PowerPoint slide images across 8 languages(French, Korean, Japanese, Spanish, German, Italian, Portuguese and Russian). This dataset includes multiple scenes, different photographic angles & distances, different light conditions. For annotation, each text line was annotated with quadrilateral bounding boxs and transcribed. The dataset can be used for tasks such as developing multilingual OCR systems.

This dataset contains handwritten text images collected from 100 individuals, including 50 Japanese, 49 Koreans and 1 Afghan. For different subjects, the corpus are different. The data diversity includes multiple cellphone models and different corpus. This dataset can be used for tasks such as handwriting OCR models, handwritten text recognition systems, and multilingual OCR pipelines

The text carrier are A4 paper, lined paper, quadrille paper, etc. The device is cellphone, the collection angle is eye-level angle. The dataset content includes Japanese composition, poetry, prose, news, stories, etc. For annotation, line-level quadrilateral bounding box annotation and transcription for the texts were annotated in the data.The dataset can be used for tasks such as Japanese OCR models and handwritten text recognition systems.

This dataset contains 4,538 Japanese handwritten text images collected from 101 individual writers, written on A4 paper. The dataset content including social livelihood, entertainment, tour, sport, movie, composition and other fields. For annotation, character-level rectangular bounding box annotation and text transcription and line-level rectangular bounding box annotation and text transcription were adopted. The dataset can be used for for training and evaluating Japanese handwriting OCR models, handwritten text recognition systems, and document understanding pipelines.

This dataset contains 105,941 images of natural scene text collected across multiple real-world environments, covering 12 languages, including 6 Asian languages and 6 European languages. The data covers multiple natural scenes, multiple photographic angles. For annotation, each image is annotated with line-level quadrilateral bounding boxes and accurate text transcriptions. The data can be used for multilingual OCR, scene text recognition, and cross-language OCR model training.

426,687 high-resolution images featuring multilingual Optical Character Recognition (OCR) data across both natural scenes and various document types. This dataset spans 20 languages, including Traditional Chinese, Simplified Chinese, Japanese, Korean, Thai, Vietnamese, Indonesian, Malay, Polish, and more. The data covers a wide range of real-world conditions—natural scenes, printed documents, handwritten notes, signs, and posters—captured from multiple countries and environments, with varied backgrounds, lighting conditions, and camera angles. All images are annotated for OCR tasks, making this dataset highly suitable for training deep learning models for text detection, recognition, and layout analysis in multi-language scenarios. The dataset complies with global data protection standards (GDPR, CCPA, PIPL), and is validated by leading AI enterprises for commercial and research applications.