Overview
Topic status: We're looking for students to study this topic.
Content-Based Multimedia Information Retrieval systems have long been developed to formulate and execute queries based on the visual or audio content and to help browsing large multimedia repositories. However, no general breakthrough has been achieved with respect to large databases with different sorts of documents. Answers to many questions with respect to speed, semantic descriptors or objective image/audio interpretations are still unanswered. Effective and efficient methods for content-based access into large scale multimedia repositories are still to be developed.
This project aims to investigate granule computing approaches for multi-media data exploration. The basic idea is to derive and use features at different levels. For example, the coarse feature descriptors can be used at the first stage to quickly screen out non-promising media data (images/audios) and the fine feature descriptors can be employed to find the truly matched images/audios later on.
Applicants with computer science, information science or engineering, especially with signal processing background, are welcome to contact the supervisor directly.
Jinglan believes that bright people can learn quickly. If you are interested but do not have the background, please contact the supervisor directly to see if there are any pathways available.
- Study level
- PhD, Masters, Honours, Vacation research experience scholarship
- Supervisors
- QUT
- Organisational unit
Science and Engineering Faculty
- Research area
- Keywords
- image searching, sound searching, multimedia browsing
- Contact
- Please contact the supervisor for enquiries.
