Key Research areas
Our Research objective is to comprehensively address failure conditions in AI-vision systems using simulation, ensuring robustness and reliability
Photorealism and Sensor design
We craft highly photorealistic environments and engineer virtual sensors that mimic real-world behaviour with high precision. This enables accurate testing of AI perception systems under a wide range of conditions, bridging the gap between simulation and reality
Edge case modelling
Our research focuses on creating rare, unexpected, and critical scenarios — the edge cases that challenge autonomous systems the most. By simulating the unpredictable, we help drive safer, smarter AI capable of handling the real world's toughest situations.
Vision perception network design and training
We develop and train advanced vision perception networks, enabling machines to see, understand, and react with human-level awareness. Our work pushes the boundaries of neural network design to improve detection, segmentation, and decision-making under complex conditions.
Immersive data generation
We generate rich, immersive datasets that capture the full complexity of real-world environments. This synthetic data powers faster, safer, and more scalable AI training, accelerating the journey from simulation to real-world deployment.