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.

Class-specific sparse codes for representing activities

In this paper we investigate the effectiveness of class specific sparse codes in the context of discriminative action classification. The bag-of-words representation is widely used in activity recognition to encode features, and although it yields state-of-the art performance with several feature descriptors it still suffers from large quantization errors and reduces the overall performance. Recently proposed sparse representation methods have been shown to effectively represent features as a linear combination of an over complete dictionary by minimizing the reconstruction error. In contrast to most of the sparse representation methods which focus on Sparse-Reconstruction based Classification (SRC), this paper focuses on a discriminative classification using a SVM by constructing class-specific sparse codes for motion and appearance separately. Experimental results demonstrates that separate motion and appearance specific sparse coefficients provide the most effective and discriminative representation for each class compared to a single class-specific sparse coefficients.

Multiple instance dictionary learning for activity representation

This paper presents an effective feature representation method in the context of activity recognition. Efficient and effective feature representation plays a crucial role not only in activity recognition, but also in a wide range of applications such as motion analysis, tracking, 3D scene understanding etc. In the context of activity recognition, local features are increasingly popular for representing videos because of their simplicity and efficiency. While they achieve state-of-the-art performance with low computational requirements, their performance is still limited for real world applications due to a lack of contextual information and models not being tailored to specific activities. We propose a new activity representation framework to address the shortcomings of the popular, but simple bag-of-words approach. In our framework, first multiple instance SVM (mi-SVM) is used to identify positive features for each action category and the k-means algorithm is used to generate a codebook. Then locality-constrained linear coding is used to encode the features into the generated codebook, followed by spatio-temporal pyramid pooling to convey the spatio-temporal statistics. Finally, an SVM is used to classify the videos. Experiments carried out on two popular datasets with varying complexity demonstrate significant performance improvement over the base-line bag-of-feature method.

Harvest: A biotextile future

- Description of the work Harvest: A biotextile future consists of four bags constructed from kombucha, each utilizing a different approach to this material. The kombucha material is a byproduct of the fermented green tea, kombucha, and is comprised of a symbiotic culture of bacteria and yeast (SCOBY) that forms a fast growing curd or pellicle on the surface of the tea. This pellicle is harvested, washed, and dried to make a material with characteristics that can range between leather and paper in handle. The pellicle is one hundred per cent cellulose, with the individual fibres growing together to produce a durable and strong non-woven textile. Techniques explored with the dry kombucha material include folding, stitching, and laser etching. The final bags were designed with reference to classic tropes of fashion accessories: the briefcase, the clutch, the valise and the handbag. The valise included three jars in which the kombucha was displayed as ‘growing’ within the bag. - Research Background This work sits within an emerging field of practice in which fashion design intersects with biotechnology. Designers such as Suzanne Lee have explored constructing garments from bacteria byproducts, and bio-artists Oron Catts and Ionat Zurr have created ‘victimless leather’ grown from cultured cells. Although still speculative, these collaborations between science and design point to new material applications for fashion. Our work contributes to this area through testing both the growing of the textile and its application to construct durable fashion artefacts. - Research Contribution Harvest: A biotextile future makes two contributions to new knowledge in the area of design for sustainability within fashion. The first contribution lies in extending the technical experimentation required to grow and manipulate the textile. For the briefcase, the pattern shape was ‘grown’ into the required shape, using a shaped container. Other techniques used in the bags included weaving, folding and laser etching the material to extend its functional and decorative properties. Experimentation with the growing and drying of the material led to the production of a wide range of physical properties, in which the material was more brittle or flexible as required. The second research contribution lies in the proposal of this material for use in durable fashion accessories. The material is still speculative and small-scale in production, however the four bags illustrate the potential for kombucha as a biodegradable alternative to leather or synthetic materials. - Research Significance This interplay of science and design research opens up an exploration for a speculative future of sustainable, biodegradable textiles using live bacteria to enable ‘homegrown’ vegan apparel. The collaborators on this project include scientist Peter Musk and fashion designers Alice Payne and Dean Brough. Harvest: A biotextile future was exhibited at the State Library of Queensland’s Asia Pacific Design Library, 1-5 November 2015, as part of The International Association of Societies of Design Research’s (IASDR) biannual design conference. The work was chosen for display by a panel of experts, based on the criteria of design innovation and contribution to new knowledge in design.

Place quality and urban competitiveness symbiosis? A position paper

Urban agglomerations—where innovation and knowledge generation activities take place—are in a tough competition to become a major player in the global knowledge economy. It is claimed that soft measures—namely quality of life and place—help in fostering and attracting talent, and consequently draw investment to these urban localities. This paper aims to scrutinise the role of soft measures in supporting urban competitiveness through a critical review of the scholarly literature. The findings shed some light on whether there is a symbiotic relationship between place quality and urban competitiveness. The paper also points out directions for future investigations.

Pulp and paper production from sugarcane bagasse

The pulp and paper industry is very large and is now well in excess of $200 billion (FAO 2009). Estimates for the amount of bagasse used in the production of pulp and paper products vary but the general consensus is that it accounts for 2–5% of global production, making it one of the highest revenue earners for the global sugarcane industry.

A germanium/single-walled carbon nanotube composite paper as a free-standing anode for lithium-ion batteries

Paper-like free-standing germanium (Ge) and single-walled carbon nanotube (SWCNT) composite anodes were synthesized by the vacuum filtration of Ge/SWCNT composites, which were prepared by a facile aqueous-based method. The samples were characterized by X-ray diffraction, field emission scanning electron microscopy, and transmission electron microscopy. Electrochemical measurements demonstrate that the Ge/SWCNT composite paper anode with the weight percentage of 32% Ge delivered a specific discharge capacity of 417 mA h g-1 after 40 cycles at a current density of 25 mA g-1, 117% higher than the pure SWCNT paper anode. The SWCNTs not only function as a flexible mechanical support for strain release, but also provide excellent electrically conducting channels, while the nanosized Ge particles contribute to improving the discharge capacity of the paper anode.

The potential benefits of divergent thinking and metacognitive skills in STEAM learning: A discussion paper

In the wake of an almost decade long economic downturn and increasing competition from developing economies, a new agenda in the Australian Government for science, technology, engineering, and mathematics (STEM) education and research has emerged as a national priority. However, to art and design educators, the pervasiveness and apparent exclusivity of STEM can be viewed as another instance of art and design education being relegated to the margins of curriculum (Greene, 1995). In the spirit of interdisciplinarity, there have been some recent calls to expand STEM education to include the arts and design, transforming STEM into STEAM in education (Maeda, 2013). As with STEM, STEAM education emphasises the connections between previously disparate disciplines, meaning that education has been conceptualised in different ways, such as focusing on the creative design thinking process that is fundamental to engineering and art (Bequette & Bequette, 2012). In this article, we discuss divergent creative design thinking process and metacognitive skills, how, and why they may enhance learning in STEM and STEAM.

Paper 1: The disruptive forces of the sharing economy

Overview: - The sharing economy unlocks a previously unused value of goods and labour, and causes disruption in established industries. - The pattern of disruption is similar regardless of the industry that's impacted. While the initial phases of disruption are transformational for many (e.g. lost jobs), often the industries end up stronger than before they were before the disruption. - Due to different in setting, upholding and enforcing standards, it is hard to assess the regulatory trade-offs. Safety, labour relations and social fairness are important factors to consider across the industry.