Indian Language Benchmark Portal

Line Recognition For Marathi Scanned Documents
Input Image
Output Image

The objective of this task is to recognize the content (i.e. text) of a Line image extracted from a page of a Marathi Scanned Documents. The right-hand side figure shows the recognized content (i.e. text) corresponding to the Line image as the input (left-hand side image).

Method Title Author(s) Hmean Recall Precision AP