You are here: Computer Science > Staff > paulp

Sakrapee (Paul) Paisitkriangkrai Postdoctoral Research Fellow The Australian Center for Visual Technologies School of Computer Science, The University of Adelaide, Australia Level 5, Innova 21, The University of Adelaide, Adelaide, SA 5006, Australia Phone: +61 8 8313 0282 Email: paulp(at)cs(dot)adelaide(dot)edu(dot)au Web: http://www.cs.adelaide.edu.au/~paulp

Research Interest

The ACVT conducts a challenging and difficult research in the area of image processing, computer vision and machine learning. Our objective is to design a new technology that can match the full capability of human vision.

Original image	The ACVT objective

Publication

Research Projects

Multi-class classification

Multi-class Boosting + Group sparsity = Feature sharing classifier

The framework is formed based on column generation based boosting. Unlike previous multi-class boosting algorithms, the new model bypasses the learning of output coding by directly learning feature coefficients for all classes. The new approach is more direct and opens up a new way of incorporating prior knowledge to multi-class problems (e.g., promoting feature sharing in this paper).

AdaBoost.ECC	Our MultiBoost (group)

Randomized boosting with matlab source code

The new model trains a single-vector parameterized classifier irrespective of the number of classes. The advantage of the new appraoch is that multi-class boosting can be trained at the same learning complexity of binary boosting.

The source code for RandomBoost can be downloaded here.

Data distribution	RandomBoost

Binary-class classification

Greedy Sparse Linear Discriminant Analysis (GSLDA)

AdaBoost	GSLDA (ours)

Online learning

Large-scale multimedia framework

Clip-based Hierarchical Representation for Near-Duplicate Video Detection

Binary-class visual detector

Pedestrian detection