The University of Southampton

Published: 9 January 2018
Illustration
Aura Vision Labs uses AI to generate high value data for retail and city planning

Pioneering researchers and students from Electronics and Computer Science (ECS) at the University of Southampton will exhibit their innovations at the Consumer Electronics Show (CES) 2018 in Las Vegas.

The world-renowned trade show will offer the globe’s first glimpse of inventions launching from ground-breaking research through demonstrations in Eureka Park, where Southampton’s Future Worlds incubator will lead the UK’s university presence at CES 2018.

University entrepreneurs will be giving live demonstrations of three featured products at CES 2018, including an artificial intelligence powered video analytics platform developed by ECS researchers.

Computer Vision expert Daniel Martinho-Corbishley, from ECS, will showcase Aura Vision Labs, a cloud-based video analytics platform that harnesses the latest innovations in artificial intelligence to generate valuable data, such as gender, age and clothing styles, for uses in high-tech retail and future smart cities. The platform, which is built on cutting-edge PhD research, uses state-of-the-art ‘deep learning’ technology to extract useful information about how people look and move from live footage captured on low-cost cameras and existing security systems.

“There are over 245 million professionally installed video surveillance cameras in operation globally, but only one per cent of this footage is currently being analysed,â€? Daniel explains. “Our technology harnesses this fantastic source of video data, working with low-cost cameras and existing security systems to provide retailers with a simple, cost-effective way to gain more accurate, in-depth and real-time information than has ever been possible before.â€?

Looking towards other applications, the international smart city revolution aims to reduce road traffic and ensure public safety through technological enhancement. The Aura Vision Labs team plans to use their technology to simultaneously track people and cars from CCTV, helping city operators identify atypical congestion and promote walking as part of a healthy lifestyle.

The University of Southampton is leading the UK’s university presence at CES 2018, with other live demonstrations on its incubator stand.

Travis Ralph-Donaldson from the Web Science Institute will bring a hands-on demonstration of Handy Kanji, an iOS app that gamifies the teaching of the Japanese writing system. The application uses intelligent stroke recognition and scoring algorithms to teach hundreds of Japanese Kanji characters.

From the Faculty of Engineering and the Environment, Associate Professor Filippo Fazi and Research Fellow Dr Marcos Simón will demonstrate Soton Audio Labs, a groundbreaking system that creates immersive 3D audio from a single soundbar. The technology uses image processing with a built-in camera to track the location of the viewer in the room, and delivers a perfect 3D experience using destructive and constructive sound cancellation techniques.

Future Worlds is the on-campus startup incubator at the University of Southampton. Since its launch in 2015, Future Worlds has been helping entrepreneurs launch groundbreaking startups with origins ranging from cutting-edge collaborative research through to student projects.


Articles that may also interest you

Share this article FacebookTwitterWeibo

Published: 8 January 2018
Illustration
Professor Lie-Liang Yang

Collaborative research between university experts from Southampton and Surrey is helping develop connectivity for future wireless systems that will combine tens of billions of machine-driven devices.

The three-year project, funded by the Engineering and Physical Sciences Research Council, is planning for Massive Machine-Type Communications (mMTC) that experts say will be critical to the success of fifth generation wireless systems (5G).

Billions of smart wireless devices are predicted in future services for smart homes, cities, transport, healthcare and environments, creating a need for scalable and efficient connectivity that could not be supported by current systems that have been primarily designed for human-initiated mobile communications.

Researchers from the Southampton Wireless group are joining with the 5G Innovation Centre (5GIC) at the University of Surrey for the project known as ‘New Air Interface Techniques for Future Massive Machine Communications’. Commercial applications will be exploited during the programme through partnerships with experienced academic and industrial partners. The project proposes to meet challenges of future mMTC by investigating and designing novel non-orthogonal multiple access, flexible duplexing, and adaptive coherent-noncoherent transmission schemes, as well as new waveforms that are tailored for the future mMTC systems.

Professor Lie-Liang Yang, from Southampton’s department of Electronics and Computer Science, says, “Our vision for the future is that everything will be smart: smart homes and smart cities connected by smart transportation systems, smart healthcare, smart environments, and much more. Achieving these objectives relies on a huge number of smart devices or machines acting in different roles, and most of them connected via the Internet of Things (IoT). Explicitly, most legacy designs for human-initiated mobile communication are not efficient for these device-centric IoT systems. With this project, we expect to investigate and design novel physical-layer and MAC techniques in order to meet the challenges. We hope our research results can be beneficial to vertical industries, subscribers, mobile operators and above all to the UK’s economic competitiveness and academic standing.”

Key technologies developed in the process will be prototyped and integrated into test bed facilities in Surrey, allowing researchers to demonstrate the viability of new design approaches and accelerate toward commercialisation. Project partners include Sony Europe, Huawei, General Electric, NEC, Thales and University spinout AccelerComm.

Southampton Wireless (SW) has been advancing wireless communications since the early 1980s and is led by Professor Lajos Hanzo. Research themes in the group include wireless communication and mobile networking, signal processing, visible-light communication, quantum communication and molecular communication.

Articles that may also interest you

Share this article FacebookTwitterWeibo

Date:
2018-2021
Themes:
Data Science / Big Data, Intelligent Systems and Machine Learning
Funding:
H2020

Researchers from across Europe will use big data to investigate population health prediction and management in this three years funded project. Big Data for Medical Analytics, or BigMedilytics, will channel funding from the EU’s Horizon 2020 programme to improve practices of life-saving care across the European continent. The University of Southampton will lead the UK pilot, alongside a range of academic and industrial partners in a part of the study that will improve the care of patients living with asthma and Chronic Obstructive Pulmonary Diseases (COPD). Specific persistent weather conditions and atmospheric pollution could be significant to the triggering of severe COPD and asthma exacerbation events in patients on regional and national scales. Therefore, we aim at developing robust and highly performing predictors for these exacerbation events using advanced big data analytics. This will be a breakthrough once achieved as we strongly believe that the prediction of these events at patient level will revolutionise care and contribute in the emerging 21st century concept of personalised medicine. The delivery of exacerbation event alerts with specific patient’s context information to clinicians, community nursing teams and pharmaceutical companies will save lives and reduce the cost of care across hospitals in the UK.

The particular challenge in the COPD and ASTHMA UK pilot in Big Medilytics is an outstanding big data problem, since it firstly requires automatic access to large volumes of data from heterogeneous data sources such as weather, atmospheric pollution, patients primary and secondary care data together with contextual information on patients’ lifestyles. Once acquired this big data needs to be intelligently mined, aggregated and organised for investigation while we develop deep machine learning algorithms to understand and detect the triggering agents of COPD or asthma exacerbation events in individual patients. The BigMedilytics project is a collaboration between the IT Innovation Centre, ECS and UK partners My mHealth and AstraZeneca UK together with NHS partners. AstraZeneca is a global, science-led biopharmaceutical business while My mHealth is a leading digital health SME that is currently the only endorsed app provider to the NHS. BigMedilytics will officially start in January 2018 and has a grant value of 600,000 euros for the University of Southampton. BigMedilytics is part of a larger partnership in Europe that includes Philips Electronics Research, the Technical University of Eindhoven in the Netherlands, the Royal College of Surgeons in Ireland, Valencia General Hospital Foundation for Clinical Research in Spain, Vienna University of Medicine in Austria.

Primary investigator

  • zas

Partners

  • Phillips Research (Netherlands)
  • MyMHealth limited (UK)
  • Astra Zeneca UK

Associated research group

  • IT Innovation Centre
Share this project FacebookTwitterWeibo

Pages