SARFusion is proposed, a robust 3D object detector that decouples object-query decoding into three parallel reasoning branches: a camera branch, a LiDAR branch, and a camera-LiDAR fusion branch that alleviates harmful cross-modal interference while preserving the benefits of multimodal fusion when complementary cues are trustworthy.
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.
Multimodal perception integrating light detection and ranging (LiDAR) and cameras has become a key paradigm for 3-D object detection, as it leverages both geometric structure and semantic information. However, in real-world autonomous driving scenarios, calibration errors, adverse weather, and sensor degradation can in...
Hui-Lin Huang, Yan Bai, Peng-Yuan Wang et al.· IEEE Sensors Journal· 0 citations
A reliability-gated cross-modal fusion model, RG- CMF, is proposed for complex-scene 3D object recognition that integrates calibration-aware RGB–LiDAR alignment, dual- branch feature extraction, reliability-gated fusion, and geometry–semantic consistency optimization and suggests that explicit reliability modeling can...
Zhongxing Duan· International Conference on...· 0 citations
Reliable 3D object detection is essential for autonomous-driving perception in complex traffic environments. Camera–LiDAR fusion combines dense visual semantics with precise point-cloud geometry to improve detection robustness. BEVFusion is a competitive bird’s-eye-view (BEV) fusion framework, but its high-capacity ima...
Zhi-Bin Shen, Cheng-Xi Huang, Guang-Lei Huo et al.· 2026 6th International Confe...· 0 citations
This work introduces SAM-AD, a domain-specific pretraining strategy that fine-tunes SAM on autonomous-driving imagery to extract feature representations with rich semantic information, and develops the Depth-Guided Wavelet Attention (DGWA) module, which suppresses high-frequency sensor noise while preserving critical c...
Zi-Ying Song, Lin Liu, Hong-Yu Pan et al.· arXiv.org· 0 citations
LiDAR–camera fusion has become a widely adopted paradigm for 3D object detection in autonomous driving by effectively leveraging the complementary geometric and semantic information provided by the two sensing modalities. However, most existing fusion methods primarily emphasize cross-modal fusion strategies while rely...
Hai-Yang Jiang, Yang Yang, Nai-Biao Cai et al.· Electronics· 0 citations
Known for his clear and elegant writing style, Bertsekas shaped fields from control and optimization to large-scale computation and artificial intelligence.