REMODEL. WP4. Vision Based Perception. T4_3. Cable Detection And Tracking. Electric Wires Dataset. Training and Test sets for Image Segmentation. v0

Zanella, Riccardo ; Caporali, Alessio ; Tadaka, Kalyan ; De Gregorio, Daniele ; Palli, Gianluca (2020) REMODEL. WP4. Vision Based Perception. T4_3. Cable Detection And Tracking. Electric Wires Dataset. Training and Test sets for Image Segmentation. v0. University of Bologna. DOI 10.6092/unibo/amsacta/6654. [Dataset]
Full text available as:
[thumbnail of REMODEL. WP4. Vision Based Perception. T4_3. Cable Detection And Tracking. Electric Wires Dataset. Training and Test sets for Image Segmentation. v0] Archive (REMODEL. WP4. Vision Based Perception. T4_3. Cable Detection And Tracking. Electric Wires Dataset. Training and Test sets for Image Segmentation. v0)
License: Creative Commons: Attribution 4.0 (CC BY 4.0)

Download (39GB)
[thumbnail of README]
Preview
Text(pdf) (README)
License: Creative Commons: Attribution 4.0 (CC BY 4.0)

Download (170kB) | Preview

Abstract

The dataset contains data for semantic segmentation of electric wires with domain independence, generated in the framework of REMODEL project. The dataset is automatically generated using chroma-key technique and contains 57300 images (where 28650 are RGB images and the other 28650 are the corresponding ground truth binary masks).

Abstract
Document type
Dataset
Creators
CreatorsAffiliationORCID
Zanella, RiccardoUniversity of Bologna
Caporali, AlessioUniversity of Bologna
Tadaka, KalyanUniversity of Bologna
De Gregorio, DanieleEyecan.ai Srl
Palli, GianlucaUniversity of Bologna
Keywords
electric wires, image segmentation, domain randomization, labeling process
Subjects
DOI
Contributors
Name
Affiliation
Type
Zanella, Riccardo
University of Bologna
Contact person
Deposit date
27 Apr 2021 10:39
Last modified
27 Apr 2021 15:14
Related identifier
Related identifier type
Relation type
Code
Project name
REMODEL - Robotic tEchnologies for the Manipulation of cOmplex DeformablE Linear objects
Funding program
EC - H2020
URI

Other metadata

Downloads

Downloads

Staff only: View the document

^