发表机构
University of Bologna; University of Parma; Stanford University; VisLab srl; Ambarella Inc.（博洛尼亚大学; 帕尔马大学; 斯坦福大学; VisLab有限公司; 安霸公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
CalibBEV是一种激光雷达-相机标定方法，通过两步BEV对齐结合CLIP对比损失实现跨模态统一，在KITTI、nuScenes基准上大幅降低了标定误差，达到最优性能。
AI 中文摘要
我们提出了CalibBEV，一种用于激光雷达-相机标定的新型鸟瞰图（Bird's Eye View, BEV）对齐方法。该方法将激光雷达与相机数据统一到共享的3D空间表示中，实现了精准且鲁棒的跨模态标定。CalibBEV利用各模态的专用架构提取传感器专属的BEV特征，并通过两步对齐过程估计标定矩阵。第一步，我们通过直接从BEV特征回归得到粗略标定矩阵来执行隐式对齐；为简化该对齐过程，我们借鉴CLIP的对比损失，强制各模态BEV表示之间的语义一致性，引导两个网络向统一特征空间收敛。第二步，我们利用BEV公式显式对齐两个模态的特征，将初始粗略估计细化为最终更精准的标定矩阵。CalibBEV的性能显著优于现有点-像素匹配方法，达到了当前最优的标定精度；在KITTI和nuScenes基准上，相比先前方法，该方法将相对旋转误差（Relative Rotation Error, RRE）分别降低51%和68%，相对平移误差（Relative Translation Error, RTE）分别降低80%和91%。
英文摘要
We present CalibBEV, a novel Bird's Eye View (BEV) alignment approach for LiDAR-camera calibration. Our method unifies LiDAR and camera data into a shared 3D spatial representation, enabling accurate and robust cross-modal calibration. CalibBEV extracts sensor-wise BEV features from each modality using domain-specific architectures and estimates the calibration matrix through a two-step alignment process. First, we perform an implicit alignment by regressing a coarse calibration matrix directly from the BEV features. To ease this alignment, we enforce semantic consistency between BEV representations across modalities using a contrastive loss inspired by CLIP, guiding both networks toward a unified feature space. In the second step, we leverage our BEV formulation to explicitly align the features of one modality with the other, refining the initial coarse estimate into a final, more accurate calibration matrix. CalibBEV significantly outperforms prior point-to-pixel matching methods, achieving state-of-the-art calibration accuracy. On the KITTI and nuScenes benchmarks, our method reduces the Relative Rotation Error (RRE) by 51% and 68%, and the Relative Translation Error (RTE) by 80% and 91%, respectively, compared to previous methods.
Journal refProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2026. p. 4345-4354
DOI：10.1109/WACV61042.2026.00423
