[Project-ideas] GSoC 2014 : Improving information retrieval methods for OCR data sets consisting of Indic scripts

[Ankit Agrawal]

Hi,

       I am Ankit Agrawal, a 4th year Dual Degree student in Electrical
Engineering with Master's specialization in Signal Processing at IIT
Bombay. I am interested in applying for GSOC 2014 under Ankur India and am
interested in the project idea - 'Improving information retrieval methods
for OCR data sets consisting of Indic scripts' because of my interests in
Computer Vision and Information Retrieval.

       I have completed coursework in Machine Learning, Natural Language
Processing, Computer Vision, Speech Processing, Information Retrieval etc.
I also participated in GSOC 2013 for scikit-image, an image processing
library in Python where I implemented some Feature Detectors and Binary
Feature Descriptors. My open-source contributions and repos of some of my
relevant course projects can be found on my
github<https://github.com/ankit-maverick>profile, the Information
Retrieval project repo specifically can be found
here <https://github.com/ankit-maverick/InformationRetrievalassignment>.

       Is the knowledge of Bengali language a prerequisite for this
project? I went through Abhishek's
InfoRescue<https://github.com/knoxxs/InfoRescue>repository and his
GSoC
blog <https://inforescue.wordpress.com/> and would like to discuss more
about the project to understand what needs to be achieved(either on the
mailing list or IRC with nick #ankit_agrawal). Thanks.

Regards,
Ankit Agrawal,
Communication and Signal Processing,
IIT Bombay.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ankur.org.in/pipermail/project-ideas-ankur.org.in/attachments/20140223/18e952e5/attachment-0002.htm>