SB-BEVFusion: Enhancing the Robustness against Sensor Malfunction and Corruptions
It addresses the vulnerability of multimodal fusion methods to missing or corrupted sensor data, which is critical for safety in autonomous driving.
The paper proposes a framework-agnostic fusion module for camera and LiDAR data that improves robustness against sensor malfunction and corruptions in 3D object detection for autonomous vehicles. The module achieves state-of-the-art performance on the MultiCorrupt dataset under various sensor deterioration scenarios, including extreme weather and sensor failure.
Multimodal sensor fusion has demonstrated remarkable performance improvements over unimodal approaches in 3D object detection for autonomous vehicles. Typically, existing methods transform multimodal data from independent sensors, such as camera and LiDAR, into a unified bird's-eye view (BEV) representation for fusion. Although effective in ideal conditions, this strategy suffers from substantial performance deterioration when camera or LiDAR data are missing, corrupted, or noisy. To address this vulnerability, we develop a framework-agnostic fusion module for camera and LiDAR data that allows for handling cases when one of the two modalities is missing or corrupted. To demonstrate the effectiveness of our module, we instantiate it in BEVFusion [1], a well-established framework to combine camera and LiDAR data for 3D object detection. By means of quantitative experiments on the MultiCorrupt dataset, we demonstrate that our module achieves favorable performance improvements under scenarios of missing and corrupted modalities, substantially outperforming existing unified representation approaches across a wide range of sensor deterioration scenarios and reaching state-of-the-art performance in scenarios of corrupted modality due to extreme weather conditions and sensor failure.