VIVA Sign Detecion Challenge 2015: CALL FOR PARTICIPATION

headerAims and Scope: Detecting traffic signs is an integral part of any intelligent car, whether it is fully autonomous or contains Advanced Driver Assistance Systems. Traffic sign detection has seen a good amount of research lately, but mainly focused on European signs. The VIVA Sign Detection Challenge will put focus on detecting US traffic signs. This challenge introduces an extension to the world’s largest US traffic sign dataset, LISA-TS, which will be used to benchmark participants.

The Challenge: The challenge is to detect four superclasses of US traffic signs: warning signs, stop signs, no turn signs, and, most challenging, speed limit signs. A training set of more than 10,000 annotated images is supplied, and the winning entry will be the one which performs the best across all four superclasses, as evaluated on a test set of 2300 pictures. Evaluation scripts are provided for easy fine-tuning.

Submission: Participants are encouraged to submit detection results for all four superclasses. The winning entry will be the one with the highest mean AUC for PR-curves across the four superclasses. The results of the VIVA Sign Detection Challenge will be presented at the Intelligent Vehicle Symposium 2015 Workshop to be held in Seoul, Korea. Participants should submit an extended abstract and accepted abstracts will be asked to present a poster at the workshop.

Participation: If you are interested and planning to participate in the VIVA Sign Detection Challenge, please send the organizers an email at viva.trafficsigns@gmail.com and visit the Sign Detection page for further details.