Student research opportunities

Real-time Road Scene Understanding

Project Code: CECS_942

This project is available at the following levels:
Engn4200, Engn R&D, Honours, Summer Scholar, Masters, PhD

Keywords:

Computer Vision, Machine Learning, Scene Understanding

Supervisor:

Dr Jose M. Alvarez

Outline:

This project aims to estimate the geometry of a road scene in real-time. The core of the algorithm is a convolutional neural network capable of classifying image pixels as horizontal,vertical or sky areas.
Current implementations can not reach real-time

Goals of this project

The goal of the project is modifying the testing algorithm to reach real-time operation (20 fps) without loosing performance.

Requirements/Prerequisites

Basic computer vision and machine learning knowledge.
MATLAB and c/c++ programming experience

Student Gain

Knowledge of computer vision and latest machine learning techniques with deep learning algorithms. Potential research opportunities at postgraduate level.