KITTI-depth

The depth completion and depth prediction evaluation are related to our work published in Sparsity Invariant CNNs (THREEDV 2017). KITTI-depth contains over 93 thousand depth maps with corresponding raw LiDaR scans and RGB images. Given the large amount of training data, this dataset shall allow a training of complex deep learning models for the tasks of depth completion and single image depth prediction. Also, we provide manually selected images with unpublished depth maps to serve as a benchmark for those two challenging tasks. We equipped a standard station wagon with two high-resolution color and grayscale video cameras. Accurate ground truth is provided by a Velodyne laser scanner and a GPS localization system. Our datsets are captured by driving around the mid-size city of Karlsruhe, in rural areas and on highways. Up to 15 cars and 30 pedestrians are visible per image.

本研究涉及的深度补全与深度预测评估相关工作，已发表于论文《稀疏不变卷积神经网络（Sparsity Invariant CNNs）》（THREEDV 2017）。KITTI深度数据集（KITTI-depth）包含超过9.3万张深度图，以及对应的原始激光雷达（LiDAR）扫描数据与RGB图像。鉴于该数据集拥有海量训练样本，其可支持针对深度补全与单图像深度预测任务的复杂深度学习模型训练。此外，我们还提供了人工遴选的、附带未公开深度图的图像集，作为上述两项极具挑战性任务的基准测试集。我们在一台标准旅行车上搭载了两台高分辨率彩色与灰度摄像机。精准的真值标签由Velodyne激光扫描仪与GPS定位系统提供。本数据集采集自德国卡尔斯鲁厄中型城市、郊区及高速公路路段。每张图像中最多可出现15辆汽车与30名行人。

搜集汇总

背景与挑战

背景概述

KITTI-depth是一个用于深度补全和单图像深度预测评估的数据集，包含超过93,000个深度图、对应的原始LiDaR扫描和RGB图像，数据采集于德国卡尔斯鲁厄的城市、乡村和高速公路场景，涉及车辆和行人，旨在支持复杂深度学习模型的训练。