University of Wollongong

About: University of Wollongong is a education organization based out in Wollongong, New South Wales, Australia. It is known for research contribution in the topics: Population & Context (language use). The organization has 15674 authors who have published 46658 publications receiving 1197471 citations. The organization is also known as: UOW & Wollongong University.

Revisiting Local Descriptor Based Image-To-Class Measure for Few-Shot Learning

Abstract: Few-shot learning in image classification aims to learn a classifier to classify images when only few training examples are available for each class. Recent work has achieved promising classification performance, where an image-level feature based measure is usually used. In this paper, we argue that a measure at such a level may not be effective enough in light of the scarcity of examples in few-shot learning. Instead, we think a local descriptor based image-to-class measure should be taken, inspired by its surprising success in the heydays of local invariant features. Specifically, building upon the recent episodic training mechanism, we propose a Deep Nearest Neighbor Neural Network (DN4 in short) and train it in an end-to-end manner. Its key difference from the literature is the replacement of the image-level feature based measure in the final layer by a local descriptor based image-to-class measure. This measure is conducted online via a k-nearest neighbor search over the deep local descriptors of convolutional feature maps. The proposed DN4 not only learns the optimal deep local descriptors for the image-to-class measure, but also utilizes the higher efficiency of such a measure in the case of example scarcity, thanks to the exchangeability of visual patterns across the images in the same class. Our work leads to a simple, effective, and computationally efficient framework for few-shot learning. Experimental study on benchmark datasets consistently shows its superiority over the related state-of-the-art, with the largest absolute improvement of 17% over the next best. The source code can be available from https://github.com/WenbinLee/DN4.git.

Rapid emergence of life shown by discovery of 3,700-million-year-old microbial structures

Abstract: Stromatolites are sedimentary formations created by the layered growth of microorganisms in shallow marine settings. Fossil stromatolites constitute some of the earliest evidence for life on Earth. Allen Nutman et al. describe metamorphosed stromatolites deposited around 3,700 million years ago in what is now Greenland. This is more than 200 million years older than the previous record-holders for earliest-known fossils, so these stromatolites rank as the Earth's earliest fossils by some margin. Although there is indirect evidence from isotope geochemistry that the pedigree of life on Earth is even older, this report is likely to be controversial.

Review: Recent progress on programming methods for industrial robots

Abstract: Although an automated flexible production cell is an intriguing prospect for small to median enterprises (SMEs) in current global market conditions, the complexity of programming remains one of the major hurdles preventing automation using industrial robots for SMEs. This paper provides a comprehensive review of the recent research progresses on the programming methods for industrial robots, including online programming, offline programming (OLP), and programming using Augmented Reality (AR). With the development of more powerful 3D CAD/PLM software, computer vision, sensor technology, etc. new programming methods suitable for SMEs are expected to grow in years to come.

A 100,000-Year-Old Ochre-Processing Workshop at Blombos Cave, South Africa

Abstract: The conceptual ability to source, combine, and store substances that enhance technology or social practices represents a benchmark in the evolution of complex human cognition. Excavations in 2008 at Blombos Cave, South Africa, revealed a processing workshop where a liquefied ochre-rich mixture was produced and stored in two Haliotis midae (abalone) shells 100,000 years ago. Ochre, bone, charcoal, grindstones, and hammerstones form a composite part of this production toolkit. The application of the mixture is unknown, but possibilities include decoration and skin protection.

The future of Blue Carbon science

Abstract: The term Blue Carbon (BC) was first coined a decade ago to describe the disproportionately large contribution of coastal vegetated ecosystems to global carbon sequestration. The role of BC in climate change mitigation and adaptation has now reached international prominence. To help prioritise future research, we assembled leading experts in the field to agree upon the top-ten pending questions in BC science. Understanding how climate change affects carbon accumulation in mature BC ecosystems and during their restoration was a high priority. Controversial questions included the role of carbonate and macroalgae in BC cycling, and the degree to which greenhouse gases are released following disturbance of BC ecosystems. Scientists seek improved precision of the extent of BC ecosystems; techniques to determine BC provenance; understanding of the factors that influence sequestration in BC ecosystems, with the corresponding value of BC; and the management actions that are effective in enhancing this value. Overall this overview provides a comprehensive road map for the coming decades on future research in BC science.