4 datasets found

Tags: Text Detection

Filter Results

  • Multi-Lingual

    The Multi-Lingual dataset is a benchmark for scene text recognition, containing 9,000 images with 9 languages.

  • CTW1500

    The dataset used for testing the proposed unsupervised pre-training method for query-based end-to-end instance segmentation (QEIS) models.

  • Total-Text

    Total-Text is a dataset for word-level arbitrary-shaped English text detection, containing 1,255 images for training and 300 images for testing.

  • CUTE80

    The dataset contains images of text in natural scenes, including street signs, logos, and product labels.
You can also access this registry using the API (see API Docs).