AFPILD: Acoustic footstep dataset collected using one microphone array and LiDAR sensor for person identification and localization

Shichao Wu, Shouwang Huang, Zicheng Liu, Qianyi Zhang, Jingtai Liu

Abstract: Highlights•Build the AFPILD by using one microphone array and one LiDAR sensor.•It contains 40 subjects with clothing and shoe type covariates considered.•Automatically annotate the footsteps with the synchronously recorded point clouds.•Elaborately conduct baselines concerning multiple covariates with CRNN.•Clothing and shoe type degrade the identification, while not on localization.