DAVIS Data Capture System

Introduction

Event-based sensors encode visual information asynchronously with low latency and high temporal resolution.
Event-based datasets are scarce, so user-friendly methods for creating said datasets are required.

This repository contributes with code to record a dataset with a DAVIS240C event camera.
The code was used to record and process the Event-based Dataset of Assembly Tasks (EDAT24).

All data are captured in raw form (.aedat) and can be processed into numpy arrays (.npy) for ease of use.

Requirements

• A DAVIS240C event camera - to obtain the data
• The jAER open-source software - to display and record the data
• An Arduino board - to trigger the commands to start and end the recordings

A detailed explanation on how to utilize the code is provided below

video.mp4

Cite our paper

If you've found this work useful for your research, please cite our paper as follows

@article{Duarte2024,
         title = {Event-based dataset for the detection and classification of manufacturing assembly tasks},
         author = {Laura Duarte and Pedro Neto},
         journal = {Data in Brief},
         volume = {54},
         year = {2024},
         doi = {https://doi.org/10.1016/j.dib.2024.110340}
}