# Native-Domain Cross-Attention for Camera-LiDAR Extrinsic Calibration Under Large Initial Perturbations

arXiv: https://arxiv.org/abs/2603.29414 · HTML: https://mlsanity.com/paper/2603.29414/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 52 references that are in the map.
- First posted on arXiv 31 Mar 2026.

## Builds on (most cited first, 26 of 26)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learning](https://mlsanity.com/paper/1702.03118/) (2017)
- [Patch n' Pack: NaViT, a Vision Transformer for any Aspect Ratio and Resolution](https://mlsanity.com/paper/2307.06304/) (2023)
- [PointGPT: Auto-regressively Generative Pre-training from Point Clouds](https://mlsanity.com/paper/2305.11487/) (2023)
- [Virtual Sparse Convolution for Multimodal 3D Object Detection](https://mlsanity.com/paper/2303.02314/) (2023)
- [Robust Multi-Modality Multi-Object Tracking](https://mlsanity.com/paper/1909.03850/) (2019)
- [CamLiFlow: Bidirectional Camera-LiDAR Fusion for Joint Optical Flow and Scene Flow Estimation](https://mlsanity.com/paper/2111.10502/) (2021)
- [R3LIVE: A Robust, Real-time, RGB-colored, LiDAR-Inertial-Visual tightly-coupled state Estimation and mapping package](https://mlsanity.com/paper/2109.07982/) (2021)
- [Pixel-level Extrinsic Self Calibration of High Resolution LiDAR and Camera in Targetless Environments](https://mlsanity.com/paper/2103.01627/) (2021)
- [CorrI2P: Deep Image-to-Point Cloud Registration via Dense Correspondence](https://mlsanity.com/paper/2207.05483/) (2022)
- [General, Single-shot, Target-less, and Automatic LiDAR-Camera Extrinsic Calibration Toolbox](https://mlsanity.com/paper/2302.05094/) (2023)
- [MOISST: Multimodal Optimization of Implicit Scene for SpatioTemporal calibration](https://mlsanity.com/paper/2303.03056/) (2023)
- [Automatic Extrinsic Calibration for Lidar-Stereo Vehicle Sensor Setups](https://mlsanity.com/paper/1705.04085/) (2017)
- [Bring Event into RGB and LiDAR: Hierarchical Visual-Motion Fusion for Scene Flow](https://mlsanity.com/paper/2403.07432/) (2024)
- [CamVox: A Low-cost and Accurate Lidar-assisted Visual SLAM System](https://mlsanity.com/paper/2011.11357/) (2020)
- [SOAC: Spatio-Temporal Overlap-Aware Multi-Sensor Calibration using Neural Radiance Fields](https://mlsanity.com/paper/2311.15803/) (2023)
- [CoFiI2P: Coarse-to-Fine Correspondences for Image-to-Point Cloud Registration](https://mlsanity.com/paper/2309.14660/) (2023)
- [3DGS-Calib: 3D Gaussian Splatting for Multimodal SpatioTemporal Calibration](https://mlsanity.com/paper/2403.11577/) (2024)
- [Online,Target-Free LiDAR-Camera Extrinsic Calibration via Cross-Modal Mask Matching](https://mlsanity.com/paper/2404.18083/) (2024)
- [Calib-Anything: Zero-training LiDAR-Camera Extrinsic Calibration Method Using Segment Anything](https://mlsanity.com/paper/2306.02656/) (2023)
- [CalibFormer: A Transformer-based Automatic LiDAR-Camera Calibration Network](https://mlsanity.com/paper/2311.15241/) (2023)

## Abstract (arXiv)

Accurate camera-LiDAR fusion relies on precise extrinsic calibration, which fundamentally depends on establishing reliable cross-modal correspondences under potentially large misalignments. Existing learning-based methods typically project LiDAR points into depth maps for feature fusion, which distorts 3D geometry and degrades performance when the extrinsic initialization is far from the ground truth. To address this issue, we propose an extrinsic-aware cross-attention framework that directly aligns image patches and LiDAR point groups in their native domains. The proposed attention mechanism explicitly injects extrinsic parameter hypotheses into the correspondence modeling process, enabling geometry-consistent cross-modal interaction without relying on projected 2D depth maps. Extensive experiments on the KITTI and nuScenes benchmarks demonstrate that our method consistently outperforms state-of-the-art approaches in both accuracy and robustness. Under large extrinsic perturbations, our approach achieves accurate calibration in 88% of KITTI cases and 99% of nuScenes cases, substantially surpassing the second-best baseline. We have open sourced our code on https://github.com/gitouni/ProjFusion to benefit the community.