Skip to main content

Projects

Current Projects


Past Projects


Object Tracking to Improve Weightlifting Technique

In this project, we use computer vision to track and examine the trajectories of weightlifters in order to help them improve their form and technique, with the goal of improving how efficiently they can carry out lifts with heavier weights whilst also avoiding injury.

Smart Auto Re-colouring Tool for Colour Blindness Suffers

In this project, we aim to develop an aid tool for suffers from colour blindness which is integrated with their current workflow and existing technology. It should be intuitive and affordable, allowing sufferers to overcome barriers to education and technology access.

Contextual Gamification of Social Interaction

In this project, we explored how game-design elements and game principles can support social interactions. We proposed a set of contextual gamification strategies, which apply flow and self-determination theory for increasing intrinsic motivation in social e-learning environments.

Generalisation in Deep Learning Abstractive Summarisation

In this project, we analysed different architectures used in abstractive text summarisation in terms of their generalisation ability by studying their behaviour with unseen data distributions and underrepresented topics. Models were implemented using OpenNMT toolkit.

Connecting Targets to Tweets: Semantic Attention-based Model for Target-specific Stance Detection

We proposed to embed a novel attention mechanism at the semantic level in the bi-directional GRU-CNN structure, which is more fine-grained than the existing token-level attention mechanism.