CalibBEV: LiDAR-Camera Calibration via BEV Alignment
2026-08-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors introduce CalibBEV, a new way to align data from LiDAR and cameras by converting both into a shared 3D bird's eye view format. Their method first makes a rough guess of how the sensors align, then fine-tunes it by matching features more closely. They use a special training technique to make sure both sensors' features look similar in this shared space. Their approach improves calibration accuracy a lot compared to older methods on popular self-driving car datasets.
LiDAR-camera calibrationBird's Eye View (BEV)3D spatial representationcontrastive lossfeature alignmentRelative Rotation ErrorRelative Translation ErrorKITTI datasetnuScenes datasetcross-modal calibration
Authors
Filippo D'Addeo, Lorenzo Cipelli, Adriano Cardace, Emanuele Ghelfi, Andrea Zinelli, Massimo Bertozzi
Abstract
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.