Event-PNP: Lightweight Event Camera-IMU Complementary Pose Solver with Cramér-Rao Lower Bound Verification

Traditional frame-based camera Perspective-n-Point (PnP) algorithms face bottleneck issues such as motion blur, limited dynamic range, and high computational complexity in challenging scenarios involving high-speed motion and drastic illumination changes. This paper proposes the Event-PNP algorithm framework, which achieves high-precision real-time six-degree-of-freedom pose estimation through tightly coupled asynchronous event streams and Inertial Measurement Unit (IMU) data at the measurement model level. The algorithm design establishes an event geometric model based on brightness constancy constraints, combining the event triggering mechanism with pixel motion models to avoid temporal resolution loss caused by traditional event frame accumulation.