CARD：面向复杂道路地形的稠密三维重建多模态自动驾驶数据集
CARD: A Multi-Modal Automotive Dataset for Dense 3D Reconstruction in Challenging Road Topography
提供准稠密多激光雷达融合深度真值（每帧约 50 万有效深度像素，约为 KITTI 深度补全的 6.5 倍），可用于训练与评测深度估计/补全模型在减速带、坑洼、非铺装路面等困难地形上的细粒度几何能力；其立体引导深度补全变体与标准化评测协议可直接作为新基准复用。
原文依据 our multi-LiDAR fusion yields 500K valid depth pixels per frame, about 6.5x more than KITTI Depth Completion
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论自动驾驶必须在多样化的路面上可靠运行，才能保障安全出行。然而，大多数驾驶数据集采集于铺设良好的平坦道路。此外，近期驾驶数据集主要为图像提供稀疏的激光雷达真值，这不足以评估深度估计与补全中的细粒度几何。为弥补这些空白，我们提出 CARD，一个多模态驾驶数据集，在富含减速带、坑洼、不规则路面与越野路段的连续序列上提供准稠密三维真值。我们的传感器套件包括同步的全局快门立体相机、前后激光雷达、来自激光雷达惯性里程计的六自由度位姿、逐车轮运动轨迹以及完整标定。值得注意的是，我们的多激光雷达融合每帧产生 50 万有效深度像素，约为 KITTI 深度补全的 6.5 倍，平均而言是其他公开驾驶数据集的 10 倍。该数据集横跨德国与意大利，覆盖 110 公里、4.7 小时。此外，CARD 提供针对道路地形不规则性的二维边界框，可为几何与感知任务提供准确基准。进一步地，我们提出针对路面不规则性的标准化评测协议，以及一个立体引导的深度补全变体，其在 CARD 上取得领先性能。我们还对最优的深度估计模型进行基准测试，以建立强基线。我们将 CARD 托管于 Hugging Face，并提供开源 SDK 与标准化划分，以支持公开排行榜与可复现评测。CARD 数据集托管于（https://huggingface.co/CARD-Data）。
英文原文摘要 1619 字符
Autonomous driving must operate reliably across diverse surfaces to enable safe mobility. However, most driving datasets are captured on well-paved flat roads. Moreover, recent driving datasets primarily provide sparse LiDAR ground truth for images, which is insufficient for assessing fine-grained geometry in depth estimation and completion. To address these gaps, we introduce CARD, a multi-modal driving dataset that delivers quasi-dense 3D ground truth across continuous sequences rich in speed bumps, potholes, irregular surfaces and off-road segments. Our sensor suite includes synchronized global-shutter stereo cameras, front and rear LiDARs, 6-DoF poses from LiDAR-inertial odometry, per-wheel motion traces, and full calibration. Notably, our multi-LiDAR fusion yields 500K valid depth pixels per frame, about 6.5x more than KITTI Depth Completion and 10x more on average than other public driving datasets. The dataset spans 110 km and 4.7 hours across Germany and Italy. In addition, CARD provides 2D bounding boxes targeting road-topography irregularities, enabling accurate benchmarking for both geometry and perception tasks. Furthermore, we introduce a standardized evaluation protocol for road surface irregularities and a stereo-guided depth completion variant that achieves leading performance on CARD. Moreover, we benchmark state-of-the-art depth estimation models to establish strong baselines. We host CARD on Hugging Face with an open source SDK and standardized splits to enable public leaderboards and reproducible evaluation. The CARD dataset is hosted on (https://huggingface.co/CARD-Data).