Research Opportunities at

Centre for Multimedia & Network Technology (CeMNet)

School of Computer Engineering

Nanyang Technological University

Singapore

All research positions in all 3 projects listed below are available immediately. Interested candidates please email a completed CV to A/Prof CHAM Tat Jen (astjcham@ntu.edu.sg, www.ntu.edu.sg/home/astjcham), indicating the project and position for which the application is made.

Project TACREA

We have vacancies for 1 postdoc, 1 research engineer and 1 PhD student to work on this defence-funded project. The project deals with localizing a user carrying/wearing a mobile camera in an urban area based on images captured from the camera. Such image-based localization technology is particularly useful in unfamiliar high-rise urban areas, where GPS triangulation may not be adequately achieved due to line-of-sight obstruction. The prior data available is a 3D geometric model layout of the urban area, which may optionally include facade appearance data. The proposed framework involves separate stages of recognizing buildings, determining image-to-model registration, and followed by pose computation.

• Postdoctoral Fellow. We are looking for candidates with a Ph.D degree from a good university, and proven research credentials (publications and/or building research systems) in a computer vision-related topic. Strong preference will be given to individuals who have track record of collaborating and leading research groups.

• Research Engineer. We are looking for candidates with a relevant Bachelors degree from a good university and a strong academic record. The candidate should have taken a subject course / module in computer vision or closely-related subject, with strong performance at undergraduate final-year project or equivalent.

• PhD Student Scholar. We are looking for candidates with a relevant Bachelors degree from a good university and a strong academic record. Preference will be given to candidates who have taken a subject course / module in computer vision or closely-related subject. The PhD candidature is expected to last four years.

SmartSpace-PROCAMS Project

We have 2 PhD student scholarships available on this project, which is a collaboration between the School of Computer Engineering, NTU and the College of Computing, Georgia Tech. The overall purpose of this project is to develop key technologies that will enable the creation of ubiquitous displays in intelligent environments, where all walls, tabletops, floors and other objects can be transformed into interactive displays -- a present-day version of the `holo-deck'. The research focus of this project is to investigate methods for projector-camera systems to be able to render high fidelity imagery on a large range of surfaces (including furniture and clothing), and under dynamically-varying conditions of occlusion and illumination.

We are looking for candidates with a relevant Bachelors degree from a good university and a strong academic record. Preference will be given to candidates who have taken a subject course / module in computer vision or closely-related subject. The PhD candidature is expected to last four years. Successful applicants will have the opportunity to spend a 6-month semester doing research at Georgia Tech with full support for travel, accommodation and allowance.

The PERCEPTER Project

We have 2 PhD student scholarships available in this broad-based project dealing with the perception and understanding of humans in video. A major goal is to achieve robustness and efficiency, not through careful engineering, but rather through principled analysis and machine learning. One research direction investigates how machine learning can be applied to tracking and detection problems that allow systems to self-learn to become more efficient (i.e. speed-up). In a significant departure from existing efforts that consider only accuracy improvements, the intention is to consider efficiency and accuracy trade-offs. In another research direction, we intend to explore machine learning approaches to creating different models for articulated human tracking that allow graceful degradation to take place in the presence of heavy occlusion, illumination variation and noise.

We are looking for candidates with a relevant Bachelors degree from a good university and a strong academic record. Preference will be given to candidates who have taken a subject course / module in computer vision or closely-related subject. The PhD candidature is expected to last four years.