Urban Semantic 3D Dataset

This dataset extends the Urban Semantic 3D (US3D) dataset developed and first released for the 2019 IEEE GRSS Data Fusion Contest (DFC19). We provide additional geographic tiles to supplement the DFC19 training data and also new data for each tile to enable training and validation of models to predict geocentric pose, defined as an object's height above ground and orientation with respect to gravity. We also add to the DFC19 data from Jacksonville, Florida and Omaha, Nebraska with new geographic tiles from Atlanta, Georgia.UPDATE: New data associated with our CVPRW'21 paper and the Overhead Geopose Challenge is now also archived here. Winning solutions from the contest, including model weights, are also now archived here. For more information about the contest, see the following links:https://www.drivendata.org/competitions/78/overhead-geopose-challengehttps://www.drivendata.co/blog/overhead-geopose-benchmarkhttps://www.drivendata.co/blog/overhead-geopose-challenge-winnershttps://github.com/drivendataorg/overhead-geopose-challengehttps://github.com/pubgeo/monocular-geocentric-pose本数据集是对2019年IEEE GRSS数据融合竞赛（DFC19）首次发布之Urban Semantic 3D（US3D）数据集的扩展。我们提供了额外的地理瓦片，以补充DFC19的训练数据，并为每个瓦片新增数据，以便训练和验证预测地心姿态的模型。地心姿态被定义为物体相对于地面的高度及其与重力方向的姿态。此外，我们还向DFC19数据中添加了来自佛罗里达州杰克逊维尔和内布拉斯加州奥马哈的新地理瓦片。更新：与我们的CVPRW'21论文及Overhead Geopose Challenge相关的新的数据也已存档在此。竞赛的获奖解决方案，包括模型权重，也已存档在此。关于竞赛的更多信息，请参阅以下链接：https://www.drivendata.org/competitions/78/overhead-geopose-challengehttps://www.drivendata.co/blog/overhead-geopose-benchmarkhttps://www.drivendata.co/blog/overhead-geopose-challenge-winnershttps://github.com/drivendataorg/overhead-geopose-challengehttps://github.com/pubgeo/monocular-geocentric-pose

搜集汇总

背景与挑战

背景概述

Urban Semantic 3D Dataset是一个扩展的城市语义3D数据集，基于2019年IEEE GRSS数据融合竞赛的数据，新增了亚特兰大等城市的地理图块和地心姿态（对象高度和方向）预测数据。数据集包含TIFF图像和点云格式，覆盖Jacksonville、Omaha和Atlanta等多个美国城市，用于机器学习、计算机视觉和遥感任务，如语义分割和立体视觉，并与多个竞赛和学术论文关联。