HP Labs India

Linguistic Resources for Handwriting Recognition

The objective of this thread of research is the development of annotated datasets of handwriting in the languages and scripts of the emerging markets, which are generally conspicuous by their absence. These are a prerequisite for all the following aspects of building handwriting recognition engines for a given script :

• fostering research in handwriting recognition by overcoming an important barrier to research
• analysis of writing styles to devise handwriting recognition strategies
• training of robust handwriting recognition algorithms
• objective evaluation of handwriting recognition algorithms
• other applications including script and writer identification

• Isolated Handwritten Tamil Character Dataset hpl-tamil-iso-char
  
  
• Isolated Tamil Handwritten Character Dataset hpl-tamil-iwfhr06-train
  Subset of Dataset hpl-tamil-iso-char used as training data for IWFHR 2006 Competition.
  
  
• Isolated Tamil Handwritten Character Dataset hpl-tamil-iwfhr06-test
  Subset of Dataset hpl-tamil-iso-char used as test data for IWFHR 2006 Competition.
  

back

top