SARFusion: Scene-Aware Routing Fusion for Robust Camera-LiDAR 3D Object Detection

2026-09-24

AI Summary

This study addresses the vulnerability of existing multimodal fusion methods to sensor degradation and their lack of fine-grained adaptability. To overcome these limitations, this work proposes a scene-aware routing framework that decouples the decoding process into three independent branches for camera, LiDAR, and fusion. By integrating scene-level reliability priors with local evidence, the framework dynamically executes an adaptive query routing strategy to achieve robust integration of multi-source information. Experimental results demonstrate that the proposed method attains 72.5 mAP on the nuScenes dataset, significantly enhancing the robustness of 3D object detection in complex environments.

Abstract

Camera-LiDAR fusion has become a prevailing paradigm for 3D object detection in autonomous driving. However, existing fusion detectors often establish strong inter-modality dependencies by decoding object queries from tightly coupled multimodal representations. Under corrupted driving conditions, such dependencies make the detector vulnerable to unreliable modalities, where degraded observations may interfere with reliable modality-specific evidence and lead to suboptimal predictions. Moreover, modality reliability can vary across both global driving scenes and individual object queries, requiring adaptive fusion decisions at a finer granularity. To bridge this gap, we reformulate robust camera-LiDAR fusion as a scene-aware branch routing problem and propose SARFusion, a robust 3D object detector. Instead of producing detections from a single fused representation, SARFusion decouples object-query decoding into three parallel reasoning branches: a camera branch, a LiDAR branch, and a camera-LiDAR fusion branch. Guided by a Scene Reliability Prior estimated from the global driving context, SARFusion further incorporates object-level evidence to route each query to the most suitable branch. This query-wise routing strategy alleviates harmful cross-modal interference while preserving the benefits of multimodal fusion when complementary cues are trustworthy. On the nuScenes test set, SARFusion achieves strong performance with 72.5 mAP and 74.4 NDS. Extensive analyses demonstrate its robustness under challenging conditions, including sensor corruptions and environmental changes.

Problem

• 3D Object Detection
• Camera-LiDAR Fusion
• Robustness
• Sensor Corruption
• Adaptive Fusion

Innovation

• Scene-Aware Routing
• Camera-LiDAR Fusion
• 3D Object Detection
• Query-wise Routing
• Robustness