LEGO bricks for training classification network - Open Research Data - Bridge of Knowledge

Your account

login

Search

LEGO bricks for training classification network

wersja 1.0

Description

The data set contains images of 447 different classes of LEGO bricks used for training LEGO bricks classification network. The dataset contains two types of images: photos (10%) and renders (90%) aggregated into respective directories. Each directory (photos and renders) contains 447 directories labeled as the official brick type number. The images contains only bricks on white background, the renders were augmented randomly so that they resemble the real pictures.

Sample images are presented below. Real photos are on the left side and renders on the right side.

Illustration of the publication
Illustration of the publication
Illustration of the publication
Illustration of the publication

Dataset file

final_dataset_prepared_447_classes.tar
8.9 GB, S3 ETag cb4260871e01f77d12c07e5edd9d15fd-18, downloads: 245
The file hash is calculated from the formula
hexmd5(md5(part1)+md5(part2)+...)-{parts_count} where a single part of the file is 512 MB in size.

Example script for calculation:
https://github.com/antespi/s3md5

File details

License:
Creative Commons: by 4.0 open in new tab
CC BY
Attribution
Raw data:
Data contained in dataset was not processed.

Details

Year of publication:
2021
Verification date:
2021-06-21
Dataset language:
English
Fields of science:
  • information and communication technology (Engineering and Technology)
  • computer and information sciences (Natural sciences)
DOI:
DOI ID 10.34808/3qfs-rt94 open in new tab
Series:
Verified by:
Gdańsk University of Technology

Keywords

References

Cite as

Authors

Version this document has several versions

DOI 10.34808/wvms-q810 represents the latest version of the data.

seen 1161 times