Semantic Robot Challenge

Research Project Goal

The Semantic Robot Vision Challenge is a research competition that is designed to push the state of the art in image understanding and automatic acquisition of knowledge from large unstructured databases of images (such as those generally found on the web). You can find more information at http://www.semantic-robot-vision-challenge.org/ The goal of this project would be to develop software to pass 'Software-Only" qualification for the competition. Given the database of images collected by the robot base available of the competition web site, learn visual models on 20 objects listed as present in the dataset by retrieving similar images from the web and identify the presence of the objects in the image database.

Research Scope

Sample Data

Available on the competition web page

Tasks

• learn the models of objects from the web
• given the learned models detect the images in the database

Challenges

Research Project Status

student names here

Point of Contact

Jana Kosecka, kosecka@cs.gmu.edu

Midterm Report

not yet submitted

Final Report

not yet submitted