MSC-Bench: Benchmarking and Analyzing Multi-Sensor Corruption for Driving Perception
This addresses safety concerns in autonomous driving by benchmarking robustness against sensor corruptions, though it is incremental as it focuses on evaluation rather than new methods.
The paper tackles the problem of low robustness in multi-sensor fusion models for autonomous driving perception when sensors are corrupted or missing, by introducing MSC-Bench, a comprehensive benchmark that reveals substantial performance degradation—e.g., up to 40% drop in mAP for some models—under adverse conditions.
Multi-sensor fusion models play a crucial role in autonomous driving perception, particularly in tasks like 3D object detection and HD map construction. These models provide essential and comprehensive static environmental information for autonomous driving systems. While camera-LiDAR fusion methods have shown promising results by integrating data from both modalities, they often depend on complete sensor inputs. This reliance can lead to low robustness and potential failures when sensors are corrupted or missing, raising significant safety concerns. To tackle this challenge, we introduce the Multi-Sensor Corruption Benchmark (MSC-Bench), the first comprehensive benchmark aimed at evaluating the robustness of multi-sensor autonomous driving perception models against various sensor corruptions. Our benchmark includes 16 combinations of corruption types that disrupt both camera and LiDAR inputs, either individually or concurrently. Extensive evaluations of six 3D object detection models and four HD map construction models reveal substantial performance degradation under adverse weather conditions and sensor failures, underscoring critical safety issues. The benchmark toolkit and affiliated code and model checkpoints have been made publicly accessible.