Family tree

Artist's Statement: These suspended shipping floats symbolise the artist's grandfather's home on Keriri (Hammond Island), where the trees are decorated with floats of all colours that have washed up on the beach. Across the entire Torres Strait, these floats, often from Asia, wash ashore and become decorative objects, strung from trees and hung from island shacks. Their vivid colours, and sometimes reflective glass surfaces, play against the lush tropical setting, while their re-use reflects the innovative character of island life. This arrangement of the floats represents the artist's family tree, which he has traced back six generations to Mer (Murray Island) and Keriri. The strings of orange floats represent his immediate family and direct lineage, each member of which is named on a float, with the totem of the family painted on the base. The remaining floats trace additional ancestry and spread further back through time and space, spanning the Torres Strait from west to east.

Activity recognition using binary tree SVM

This paper presents an effective classification method based on Support Vector Machines (SVM) in the context of activity recognition. Local features that capture both spatial and temporal information in activity videos have made significant progress recently. Efficient and effective features, feature representation and classification plays a crucial role in activity recognition. For classification, SVMs are popularly used because of their simplicity and efficiency; however the common multi-class SVM approaches applied suffer from limitations including having easily confused classes and been computationally inefficient. We propose using a binary tree SVM to address the shortcomings of multi-class SVMs in activity recognition. We proposed constructing a binary tree using Gaussian Mixture Models (GMM), where activities are repeatedly allocated to subnodes until every new created node contains only one activity. Then, for each internal node a separate SVM is learned to classify activities, which significantly reduces the training time and increases the speed of testing compared to popular the `one-against-the-rest' multi-class SVM classifier. Experiments carried out on the challenging and complex Hollywood dataset demonstrates comparable performance over the baseline bag-of-features method.