Detection and recognition of traffic signs

Research Project Goal

We have a large corpus of panoramic video that is geo-referenced. The video has been collected by a car instrumented with high-recision GPS and inertial sensors. The goal of this project is to detect a subset of traffic signs (stop, yield, speed limit) encountered along the way in the front-facing view of the camera.

Research Project Scope

Design traffic sign detector, exploiting the knowledge of the traffic signs and the environment context determining where the traffic signs appear.

Data

Video stream of the images acquired by a camera mounted on a car will be available. Sample sequence can be downloaded from here

Tasks

• Learn scale invariant model of the e.g. STOP sign (color can be also used as a cue here)
• Implement efficient detection algorithm. First you can brute forces search all possible windows at all possible scales. In the second stage the knowledge where traffic signs appear in the images can be used to either reduce the search or reject the false positives.

Research Project Status

student names here

Point of Contact

Jana Kosecka

Midterm Report

not yet submitted

Final Report

not yet submitted