CTD: Composite texture dataset

Overview

Datasets used in the paper: PDF

@inproceedings{giraud2019texture,
  title={Texture-aware superpixel segmentation},
  author={Giraud, R{\'e}mi and Ta, Vinh-Thong and Papadakis, Nicolas and Berthoumieu, Yannick},
  booktitle={2019 IEEE International Conference on Image Processing (ICIP)},
  pages={1465--1469},
  year={2019}
}

• mix-Brodatz: Natural composite texture dataset using images from the Brodatz dataset

• mix-Stripes: Synthetic composite texture dataset combining lines with different orientations

Details

• Size: 11.6MB
• Platform: Img are in .png format and ground truthes are given with both .png and .mat (matlab) formats.
• Environment: No specific environment is needed to exploit the data.

License

(C) Rémi Giraud, 2020
remi.giraud@u-bordeaux.fr
https://remi-giraud.enseirb-matmeca.fr
ENSEIRB-MATMECA (Bordeaux INP), Laboratory IMS