Student research opportunities
What, Where and How Many? Describing a Supermarket
Project Code: CECS_782
This project is available at the following levels:
CS single semester, Engn4200, Engn R&D, Honours, Summer Scholar, Masters
Keywords:
computer vision, machine learning
Supervisor:
Dr Yuhang ZhangOutline:
While shopping in a supermarket, most people can see what products are placed on which shelf as well as how many copies are left, easily. Yet for people who are visually impaired, the above task can be challenging or even impossible. This project aims to develop a scene understanding system which can describe the shelves in a supermarket automatically.The expectation is that this system can work as a bionic eye for people with weak vision in a supermarket.
The project is an extension to the work in the following web page:
http://yuhang.rsise.anu.edu.au/webmarket.php
Images and some basic C programs are available. However, significant contribution is still needed to equip the system with the above capability.
Requirements/Prerequisites
The students must be self-motivated and should have above average C/C++ programming skills. Solid maths background is also desired. Knowledge in image processing and machine learning can be a plus.
Background Literature
Y. Zhang, L. Wang, R. Hartley and H. Li: Where's the Weet-Bix? ACCV2007.
Y. Zhang, L. Wang, R. Hartley and H. Li: Handling Significant Scale Difference for Object Retrieval in a Supermarket. DICTA2009.
Lubor Ladicky, Paul Sturgess, Karteek Alahari, Christopher Russell, Philip H. S. Torr: What, Where and How Many? Combining Object Detectors and CRFs. ECCV2010
