Visual Reasoning Technology
Exploring how vision AI can move beyond detection to understand context, reason about events and support intelligent decisions.
Intelligent Decision-Making Reasoning
Researching how confidence, persistence, context and operational rules can turn raw AI detections into more meaningful events.
Reusable Vision Models
Building vision foundations that can be reused and adapted across multiple applications, environments and real-world problems.
Vision-Language Intelligence
Exploring multimodal models that combine visual perception and language reasoning to understand complex scenes and operational conditions.
High-Performance AI Inference
Researching lower latency and more efficient inference through precision optimisation, quantisation and model execution techniques.
Parallel Vision Processing
Investigating how multiple video streams, people and AI workloads can be processed concurrently for scalable real-time vision.
Edge & Heterogeneous Computing
Exploring how visual workloads can be distributed across cameras, edge devices and accelerated compute for efficient real-world deployment.
Temporal & Context-Aware Intelligence
Researching how time, persistence, environmental context and event history can improve how visual AI interprets what is really happening.