RCBEVDet: A Robust Classifier for Birds-Eye-View Detection

Abstract

This paper proposes RCBEVDet, a robust multi-modal fusion framework for autonomous driving 3D object detection.

Key Contributions

1. Novel fusion architecture: Integrates multi-scale features from different sensors
2. Robust detection: Maintains high accuracy even in adverse weather conditions
3. Efficient implementation: Real-time performance on edge devices

Results