Agnostic Tomography

Agnostic tomography aims to reconstruct the shape or properties of an unknown object or quantum state without prior knowledge of its class or type. Current research focuses on developing efficient algorithms, often leveraging implicit representations like Gaussian splatting or neural radiance fields, and exploring different model architectures to handle dynamic objects and articulated structures in both classical and quantum settings. This research is significant for advancing robotics, computer vision, and quantum computing by enabling robust and generalizable perception and state estimation in complex and uncertain environments.