Annotated Chemical Apparatus Image Dataset

Published: 15 December 2023| Version 1 | DOI: 10.17632/8p2hvgdvpn.1
Contributors:
,
,

Description

This dataset consists of 5078 annotated images of chemical apparatuses. These images were obtained by extracting diverse frames from videos of chemical experiments recorded using a smartphone. Annotations cover six experimental apparatuses and the experimenter's hands. The dataset is divided into three se ts: training, validation, and test, each with accompanying annotation texts in YOLO (You Only Look Once) format.

Files

Institutions

Waseda Daigaku - Nishiwaseda Campus

Categories

Chemistry, Object Detection, Image Classification

Funding

Ministry of Education, Culture, Sports, Science and Technology

Grant-in-Aid for Transformative Research Areas (A), JP22H05380

Licence