Robots could lend a helping hand

Giraff projectWith a rapidly ageing population, the number of people with dementia and other age-related disabilities is expected to soar by 20501.

Coupled with warnings about future shortages of health workers and doctors2, it’s clear other options need to be found.

The concept of applying findings from different technological areas to assist people in their daily activities, while also alleviating pressures on health professionals and carers, has emerged as a potential solution.

It is these Quality of Life Technologies (QOLTs) that are of particular interest to Dr Oscar Martinez Mozos, Lecturer in the School of Computer Science at the University of Lincoln (UK).

During the last two and a half years Dr Mozos has been working on assistive robotic technologies at Kyushu University in Japan, where he currently keeps a position as an external collaborative researcher.

He is bringing together some of the world’s leading researchers in this area at a special session for an international conference to be held in Spain from June 10-14.

Dr Mozos, whose research specifically focusses on the application of computer science to service robotics, assistive technologies, medicine and industry, said: “Typical applications for QOLTS are support aids for people with some kind of disability such as assistance robots and rehabilitation technologies, but they also include powerful tools to improve well-being of individuals and society in general. It’s basically using technology to enhance the lives of people in any way, whether it is by programming robots to perform specific tasks or through medicine. The aim of the conference is to bring together the top academics in this field and link the various disciplines, which include engineering, computer science, medicine, psychology and social sciences.”

Dr Mozos is now working with colleagues to collate all of the research strands in a special issue of the IEEE Journal of Biomedical and Health Informatics in June 2014.

Professor Cipriano Galindo, from the University of Malaga, Spain, and Professor Adriana Tapus, from the ENSTA-ParisTech, France, are also guest editors.

A project Professor Galindo is involved in aims to develop a system that will perform a range of services, including data collection and analysis of human behaviours through a ‘telepresence’ robot. The Giraff+ system will be installed and evaluated in at least 15 homes of elderly people in Sweden, Italy and Spain.

Professor Tapus is currently working on how assistive robots can provide affordable and personalised cognitive assistance, motivation and companionship to users suffering from conditions related to ageing or Alzheimer’s disease.

She said: An important and growing trend in modern robotics research is to create robots with human-like qualities, which will allow robots to interact naturally with humans and to become a part of our lives. The main advantages of my research project are that it provides time-extended personalised cognitive and social interaction and “exercise” in a robot-supervised fashion. This is an entirely novel area of research in assistive and rehabilitation robotics and it opens up a broad avenue for future discovery and development.”

Research Presentation by Prof Ales Leonardis on Wed 29/05, 4pm

Prof Ales Leonardis (University of Birmingham) is presenting his research on “Combining compositional shape hierarchy and multi-class object taxonomy for efficient object categorisation”  in a joint Psychology and Computer Science seminar on Wed, 29th May at 4pm (in room MC1001). All students and staff are invited.

Here the abstract of Ales’ presentation:

Visual categorisation has been an area of intensive research in the vision community for several decades. Ultimately, the goal is to efficiently detect and recognize an increasing number of object classes. The problem entangles three highly interconnected issues: the internal object representation, which should compactly capture the visual variability of objects and generalize well over each class; a means for learning the representation from a set of input images with as little supervision as possible; and an effective inference algorithm that robustly matches the object representation against the image and scales favorably with the number of objects. In this talk I will present our approach which combines a learned compositional hierarchy, representing (2D) shapes of multiple object classes, and a coarse-to-fine matching scheme that exploits a taxonomy of objects to perform efficient object detection.
Our framework for learning a hierarchical compositional shape vocabulary for representing multiple object classes takes simple contour fragments and learns their frequent spatial configurations. These are recursively combined into increasingly more complex and class-specific shape compositions, each exerting a high degree of shape variability. At the top-level of the vocabulary, the compositions represent the whole shapes of the objects. The vocabulary is learned layer after layer, by gradually increasing the size of the window of analysis and reducing the spatial resolution at which the shape configurations are learned. The lower layers are learned jointly on images of all classes, whereas the higher layers of the vocabulary are learned incrementally, by presenting the algorithm with one object class after another.
However, in order for recognition systems to scale to a larger number of object categories, and achieve running times logarithmic in the number of classes, building visual class taxonomies becomes necessary. We propose an approach for speeding up recognition times of multi-class part-based object representations. The main idea is to construct a taxonomy of constellation models cascaded from coarse-to-fine resolution and use it in recognition with an efficient search strategy. The structure and the depth of the taxonomy is built automatically in a way that minimizes the number of expected computations during recognition by optimizing the cost-to-power ratio. The combination of the learned taxonomy with the compositional hierarchy of object
shape achieves efficiency both with respect to the representation of the structure of objects and in terms of the number of modeled object classes. The experimental results show that the learned multi-class object representation achieves a detection performance comparable to the current state-of-the-art flat approaches with both faster inference and shorter training times.

Games Computing commended in the Student Union Awards 2013

We are pleased to announce that Games Computing were commended in the Student Union Awards 2013 for the Best Course Award.

The recent upgrade of  facilities for this course has been greatly appreciated by students. The exemplary use of blackboard as a timely communication tool particularly stood out to the panel as a leading example of how virtual learning spaces can be used effectively.

Computer Science hosts visiting Professor from India

DSC04579Professor Atul Donsai visited the University of Lincoln sponsored by UKIERI: The UK-India Education and Research Initiative. Professor Atul lectures in Computer Science at Saurashtra University, Rajkot, India

Professor Atul used the visit as an opportunity to progress a number of collaborative research initiatives and to help build a stronger partnership between the University of Lincoln and Universities in his region. He made a number of good new friends and enjoyed exploring the vibrant and historic City of Lincoln.

Professor Atul said: Lincoln is an excellent place to study, with the opportunity to learn from experts in the field of Computer Science, whilst also enjoying a traditional English lifestyle.”

University of Lincoln, UK