Post Fusion Bird's Eye View Feature Stabilization for Robust Multimodal 3D Detection

TL;DR

This paper introduces a lightweight Post Fusion Stabilizer (PFS) module that enhances the robustness of multimodal 3D detection in autonomous driving by stabilizing features under domain shifts and sensor failures, without retraining existing models.

Contribution

The paper presents a novel, lightweight PFS module that improves robustness of BEV fusion detectors against domain shifts and sensor failures without modifying or retraining the original detection architecture.

Findings

Achieves state-of-the-art robustness on nuScenes benchmark.

Improves camera dropout robustness by +1.2% mAP.

Enhances low-light detection performance by +4.4% mAP.

Abstract

Camera-LiDAR fusion is widely used in autonomous driving to enable accurate 3D object detection. However, bird's-eye view (BEV) fusion detectors can degrade significantly under domain shift and sensor failures, limiting reliability in real-world deployment. Existing robustness approaches often require modifying the fusion architecture or retraining specialized models, making them difficult to integrate into already deployed systems. We propose a Post Fusion Stabilizer (PFS), a lightweight module that operates on intermediate BEV representations of existing detectors and produces a refined feature map for the original detection head. The design stabilizes feature statistics under domain shift, suppresses spatial regions affected by sensor degradation, and adaptively restores weakened cues through residual correction. Designed as a near-identity transformation, PFS preserves performance while improving robustness under diverse camera and LiDAR corruptions. Evaluations on the nuScenes benchmark demonstrate that PFS achieves state-of-the-art results in several failure modes, notably improving camera dropout robustness by +1.2% and low-light performance by +4.4% mAP while maintaining a lightweight footprint of only 3.3 M parameters.