Published October 20, 2025 | Version 1.0
Dataset Open

Smatable Vibration Gesture Dataset (4-Channel, 44.1 kHz, Swipe UDLR)

Creators

  • 1. ROR icon Nara Institute of Science and Technology

Contributors

Data curator:

  • 1. ROR icon Nara Institute of Science and Technology

Description

This dataset provides 4-channel table-vibration recordings (44.1 kHz, 16-bit PCM) for swipe-direction gesture recognition (Up/Down/Left/Right). Signals were captured using four synchronized piezoelectric sensors attached under tabletops in two scenarios: (i) DataByPerson (three participants on the same table), and (ii) DataByTable (one participant across three tables). Each target has ten sessions; each session contains 4 classes × 10 trials = 40 samples. The archive includes raw WAV files. See README for protocol and recommended evaluations.

Files

README.md

Files (957.8 MB)

Name Size Download all
md5:f14c2ca9c15fb078932f33e0a5d39f07
5.1 kB Preview Download
md5:9e47b9cb25dc8f094365ae64df77dd91
957.8 MB Preview Download

Additional details

Related works

Is documented by
Journal article: 10.1109/ACCESS.2023.3343500 (DOI)

Software

Repository URL
https://github.com/tianheng-ling/smatable
Programming language
Python
Development Status
Active