The CVC has already organized two ICDAR Robust Reading Competition, in 2011 and 2013.
Robust reading is a big challenge an area of a great interest among the international scientific community. Besides, the CVC is been working for a long time in this issue, becoming one of the biggest experts of this field, therefore the center decided to start those competitions.
Given the high interest demonstrated during the ICDAR 2011 Robust Reading Competition and the increasing interest of the community on complex text containers, the ICDAR 2013 Robust Reading Competition revisited the past two challenges on static real scene images and born-digital images, and introduced a brand new challenge on video data.
Since the beginning of the 2013 challenge last summer, there have been already 400 registered users when the international community is around 300 people, 50k page views and over 1500 submitted results – 51 of which public.
Even though the competitions have been a success by their own, the organizers wanted to take a step further so the CVC Annotation and Performance Evaluation Platform has been created. It is a Web service; a platform that supports the definition of comprehensive ground truth information at different text representation levels while it offers centralized management and quality control of the ground truthing effort. It implements a range of state of the art performance evaluation algorithms and offers functionality for the definition of evaluation scenarios, on-line calculation of various performance metrics and visualization of the results.