We are a group of researchers who are enthusiastic about Visual Information Learning and Analysis. We call our group VILA (geographically, a seaport in and the capital of Vanuatu). It aims to explore the synergy between computer vision problems and machine learning techniques to design better visual information analysis algorithms.
Lei Wang (Academic staff, Group Leader)
Luping Zhou (Academic staff, DECRA Fellow 2016)
Zhimin Gao (PhD student)
Yan Zhao (PhD student)
Zhongyan Zhang (PhD student)
Biting Yu (PhD student)
Melih Engin (MPhil student)
Ian Comor (Master student)
Maciej Zieba (Visiting Fellow)
Shu Chen (Visiting Fellow)
Jianjia Zhang (Data Analyst, CSIRO)
Lingqiao Liu (University of Adelaide, DECRA Fellow 2017)
Xinwang Liu (Academic staff, National University of Defence Technology, China)
Huangjing Ni (Research Fellow, Chinese Academy of Sciences)
Current research focus
Generic Image recognition
Image and object recognition has recently made significance progress with the introduction of the Bag-of-features model. Recognition accuracy has been improved to an unprecedentedly high level. Our recent research in this regard is focused on i) designing compact and discriminative codebooks to handle large-scale recognition tasks; ii) developing better coding and pooling schemes which are critical to the recognition performance; and iii) inventing new models and algorithms to address fine-grained image recognition problem. Some recent publications are listed below.
Content-based Image Retrieval
Content-based image retrieval has recently witnessed its fast development due to the pervasive use of mobile platform and the explosion of the volumes of images over the Internet. Our recent research in this regard is focused on i) image retrieval on archival photographic collections, with National Archives of Australia; ii) Fashion image retrieval for online shopping; and iii) better image retrieval models and algorithms. Some recent works are listed below.
Multiple Kernel Learning and Feature Selection.
Multiple kernel learning (MKL) provides not only an elegant framework to learn an optimal kernel for a classification task but also an effective manner to integrate heterogeneous information sources. Feature selection has become more important with the ubiquitous use of high-dimensional data. Our recent research in this regard is focused on i) incorporating the radius information into MKL; ii) MKL with missing channels and sample-based adaptivity; and iii) local and global similarity preserving feature selection. Some recent publications are listed below.
Medical Imaging Analysis.
Medical imaging analysis can provide direct, sensitive and consistent measures for the diagnosis and understanding of diseases. Our research in this regard is focused on the following fields: i) Neuroimage analysis for Alzheimer’s disease (AD), a fatal and progressive neurodegenerative disease that has caused serious socioeconomic problems. Specifically, we study the following problems for the prediction of AD: brain connectivity analysis, distinctive feature selection, and hippocampal shape analysis; ii) novel algorithms for cells image classification. Some recent works are listed below.