Recurrent Convolutional Network for Video-based Person Re-Identification

In this paper we propose a novel recurrent neural networkarchitecture for video-based person re-identification.Given the video sequence of a person, features are extracted from each frame using a convolutional neural network that incorporates a recurrent final layer, which allows information to flow between time-steps. The features from all time steps are then combined using temporal pooling to give an overall appearance feature for the complete sequence. The convolutional network, recurrent layer, and temporal pooling layer, are jointly trained to act as a feature extractor for video-based re-identification using a Siamese network architecture.Our approach makes use of colour and optical flow information in order to capture appearance and motion information which is useful for video re-identification. Experiments are conduced on the iLIDS-VID and PRID-2011 datasets to show that this approach outperforms existing methods of video-based re-identification.

https://github.com/niallmcl/Recurrent-Convolutional-Video-ReID
Project Source Code

'It's Almost Like Talking to a Person': Student Discolsure to Pedagogical Agents in Sensitive Setting

Online chatbots (also known as pedagogical agents or virtual assistants) are becoming embedded into the fabric of technology, both in educational and commercial settings. Yet understanding of these technologies is inchoate and often untheorised, influenced by individuals’ willingness to trust technologies, aesthetic appearance of the chatbot and technical literacy, among other factors. This paper draws upon data from two research studies that evaluated students’ experiences of using pedagogical agents in education using responsive evaluation. The findings suggest that emotional connections with pedagogical agents were intrinsic to the user’s sense of trust and therefore likely to affect levels of truthfulness and engagement. They also indicate that the topic of the pedagogical agent-student interaction is key to the student’s experience. The implications of these studies are that truthfulness, personalisation and emotional engagement are all vital components in using pedagogical agents to enhance online learning.

Using a whole person approach to support people with cancer: a longitudinal, mixed methods service evaluation

Introduction: Improved models of care are needed to meet all the support needs of people with cancer, which encompass psychological, emotional, physical, spiritual, sexual, occupational, social and existential needs. The aim of this paper is to (1) evaluate short and long-term impacts of using a whole person approach to support people with cancer on the Living Well with the Impact of Cancer Course (LWC); (2) use these data to inform strategic decisions about future service provision at Penny Brohn UK. Methods: Longitudinal mixed-methods service evaluation (n=135). Data collected included health related quality of life (HRQoL) (FACIT-SpEx); Concerns (types and severity - MYCaW); lifestyle behaviour (bespoke questionnaire) and participants’ experiences over 12 months post course. Results: Statistically and clinically significant improvements from baseline - 12 months in severity of MYCaW Concerns (n=64; p<0.000) and mean total HRQoL (n=66; p<0.000). The majority of MYCaW concerns were ‘psychological and emotional’ and about participants’ wellbeing. Spiritual, emotional and functional wellbeing contributed most to HRQoL improvements at 12 months. Barriers to maintaining healthy lifestyle changes included lack of support from family and friends, time constraints, and returning to work. 3-6 months post-course was identified as the time when more support was most likely to be needed. Conclusions: Using a whole person approach for the LWC enabled the needs of participants to be met, and statistically and clinically significant improvements in HRQoL and MYCaW Concerns were reported. Qualitative data analysis explored how experiencing whole person support enabled participants to make and sustain healthy lifestyle changes associated with improved survivorship. Barriers experienced to making health behaviour change were also identified. These data then informed wider and more person-centred clinical provision to increase the maintenance of positive long-term behaviour changes. Comparison of whole person approaches to cancer treatment and support and standard care are now urgently needed.

Building a person home behavior model based on data from smart house sensors 2015

Dissertação para a obtenção do grau de Mestre em Engenharia Electrotécnica Ramo de Energia

Person Localization Using Sensor Information Fusion

Nowadays the incredible grow of mobile devices market led to the need for location-aware applications. However, sometimes person location is difficult to obtain, since most of these devices only have a GPS (Global Positioning System) chip to retrieve location. In order to suppress this limitation and to provide location everywhere (even where a structured environment doesn’t exist) a wearable inertial navigation system is proposed, which is a convenient way to track people in situations where other localization systems fail. The system combines pedestrian dead reckoning with GPS, using widely available, low-cost and low-power hardware components. The system innovation is the information fusion and the use of probabilistic methods to learn persons gait behavior to correct, in real-time, the drift errors given by the sensors.