Important Dates

  • 10th Jan 2015:
    Website online
  • 10th Jan 2015:
    Registration open
  • 06th Feb 2015:
    Sample dataset available
  • 28th Feb   10th March 2015: Registration closes
  • 1st March 2015:  
    Training dataset available
  • 10th   15th March 2015: Validation dataset available
  • 20th   25th   31st March 2015: Submission of Systems
  • 25th   30th April 2015: Test dataset available

Latest News

  • 06th Feb 2015: Sample Dataset available!
  • 10th Jan 2015: Website online
  • 10th Jan 2015: Registration open!

Dataset

A dataset comprising video words for each of the ten scripts will be provided. The dataset contains at least 1000 words (Colour) for each script extracted from various sources (news, sports etc.) The dataset is divided into three parts,


  1. Training set (60%)
  2. Validation set (10%)
  3. Test Set (30%),

Sample Dataset



Train Dataset:

The link to download the training dataset will be send to the registered participants



Validation Dataset:

The link to download the validation dataset will be send to the registered participants



Test Dataset:

The link to download the Test dataset will be send to the registered participants



Dataset Usage:

The CVSI 2015 dataset contains video words of 10 different scripts/languages, namely, English, Hindi(Devnagari), Bengali(Bangla), Oriya, Gujrathi, Punjabi, Kannada, Tamil, Telegu, and Arabic.


The dataset is made publicly available ONLY for RESEACH PURPOSES. Use of dataset for commercial purposes/applications is not allowed. The use of dataset requires citing the following:


Nabin Sharma, Ranju Mandal, Rabi Sharma, Umapada Pal, and Michael Blumenstein, "ICDAR2015 Competition on Video Script Identification (CVSI 2015)", In Proc. 13th International Conference of Document Analysis and Recognition (ICDAR 2015), pp. 1196-1200, 2015, [pdf].