|
CVIT has been actively working on developing systems, algorithms, tools and resources for document image understanding with special focus on Indian scripts. It has looked into developing recognizers (both OCRs and OHRs) for recognizing printed as well as online-handwritten documents. The core engines used for recognition are learnt from a corpus. Appropriate machine learning techniques are designed to suit the specific patterns in Indian scripts. Performances are validated on large annotated corpus. In recent years, we have been working along with a team of other institutions in the country for developing language technologies required for building robust solutions.
CVIT has also contributed to the development of technologies for Digital Library of India (DLI). DLI aims at archiving 1 Million books over Internet and making them freely accessible to the public. In addition to developing document specific image processing techniques, We have also been active in developing search/access methods for large collection of document images, even when recognizers are unavailable or unreliable (for example historical documents).
Specific activities in this area include:
* Developing OCR Systems
* Algorithms for Character Recognition
* Development of large annotated corpus for training and evaluation.
* Search in digital library of document images.
* Language specific issues in preprocessing and segmentation.
* Design of classification architectures.
* Application Systems
Related Papers:
* Jyotirmoy Banerjee, Anoop M. Namboodiri and C.V. Jawahar "Contextual Restoration of Severely Degraded Document Images" in Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR 09), 20-25 June, 2009, Miami Beach, Florida, USA.
* Pramod Sankar K and C.V. Jawahar "Probabilistic Reverse Annotation for Large Scale Image Retrieval " in Proc of IEEE Computer Society Conference on Computer Vision and Pattern Recognition,Minneapolis, Minnesona, 18-23 June, 2007.
* Venkata Rasagna, Anand Kumar, C.V. Jawahar and R. Manmatha "Robust Recognition of Documents by Fusing Results of Word Clusters" in Proceedings of 10th International Conference on Document Analysis and Recognition(ICDAR 09), 26-29 July, 2009, Barcelona, Spain.
* Naveen Tewari and Anoop M. Namboodiri "Learning and Adaptation and Improving Handwritten Character Recognizers" in Proceedings of 10th International Conference on Document Analysis and Recognition(ICDAR 09), 26-29 July, 2009, Barcelona, Spain.
* A. Balasubramanian, Million Meshesha and C. V. Jawahar, "Retrieval from Document Image Collections" in Proceedings of Seventh IAPR Workshop on Document Analysis Systems, 2006 LNCS 3872), pp 1-12.
Related Thesis:
* Recognition and Retrieval from Document Image Collections
Million Meshesha, Year of Completion : 2008
* Document Annotation and Retrieval Systems
A. Balasubramanian, Year of Completion : 2006
* Word Hashing for Efficient Search In Document Image Collections
Anand Kumar, Year of Completion : 2008
Faculty
- C V Jawahar
- P J Narayanan
(Head)
|