Optical Character Recognition
OtherDocumentBounding Box
Annotation of an image and drawing bounding boxes around each word, transcribing it, and markingoverlaid/scene text; grouping word annotations into lines based on physical position, orientation, and semantic consistency, indicating if the line isoverlaid/scene text; grouping lines into meaningful semantic paragraphs, specifying if the paragraph is overlaid/scene text.
Annotation of an image and drawing bounding boxes around each word, transcribing it, and markingoverlaid/scene text; grouping word annotations into lines based on physical position, orientation, and semantic consistency, indicating if the line isoverlaid/scene text; grouping lines into meaningful semantic paragraphs, specifying if the paragraph is overlaid/scene text.
2025 - Present