PrivateMap-Bench: A Privacy-Utility Benchmark for Indoor SLAM Map Sharing
收藏DataCite Commons2026-05-07 更新2026-05-18 收录
下载链接：
https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/WFY0QT
下载链接
链接失效反馈官方服务：
资源简介：
PrivateMap-Bench is a TurtleBot4 benchmark for measuring privacy leakage and utility loss in indoor SLAM map sharing. This dataset release contains sanitized processed SLAM artifacts, filtered artifacts, aggregate privacy and utility results, generated figures and tables, annotations, configuration files, executable benchmark code,tests, and documentation for reproducing the PrivateMap-Bench NeurIPS 2026 processed-artifact benchmark.
The benchmark evaluates floor-plan leakage, room-graph leakage, trajectory leakage, routine leakage, map compactness, and A* navigation utility across 15 TurtleBot4 runs from three indoor layout variants. The release intentionally excludes raw ROS bags, raw camera streams, unreviewed RGB keyframes, contact sheets, local paths, Git history, machine-specific caches, and private identifying information.
提供机构：
Harvard Dataverse
创建时间：
2026-05-07
