Super-resolution results on document images
All the input test images and their corresponding high-resolution images can be downloaded from here. (link)
Indic Keyboards on Android
On-screen keyboards for Devanagari, Kannada, Tamil, Telugu and Malayalam scripts on Android. Users can choose from either InScript or 3x4 keyboard layout. In addition, KaGaPa layout for Kannada and TamilNet99 layout for Tamil are also supported. (link)
Multi-script Robust Reading Competition (MRRC)
MRRC is an unique competition, with multiple tasks of multi-script scene text localization, segmentation and recognition. MRRC was a part of ICDAR 2013 competitions. The data set used in MRRC is available for download from the competition site. (link)
MAST: Multi-Script Annotation toolkit for Scene Text
A semi-automatic tool for annotation of multi-script text from camera captured scene images. (link)
Five distinct scene image word datasets with pixel level annotation
3600 annotated word images at pixel level are available for download. These images may be used to train classifier. (link)
Multilingual Indic Keyboard Interface
An interface which will allow the usage of Indic scripts on both Windows and Linux, using Java and Unicode. (link)
OHR Dataset: Isolated Tamil Words
A collection of 15000 isolated Tamil words towards the development of a writer independent handwriting reocognition framework. (link)
QRS Detection: MATLAB Code
A MATLAB code that uses dynamic ploison index to extract the R-peaks from the QRS-complex of the ECG signal. (link)
Word Image Dataset for 11 Indian Scripts
This dataset contains 2,20,000 word images (of printed text) from 11 different Indian language scripts (20,000 images for each language) namely (link)
Font and Background Color Independent Text Binarization
A parameter free MATLAB code for text binarization, which handles varying text sizes and the presence of inverse text (link)
Telugu Characters Dataset