Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data

Abstract: 

Segmentation of unseen industrial parts is essential for autonomous industrial systems. However, industrial components are texture-less, reflective, and often found in cluttered and unstructured environments with heavy occlusion, which makes it more challenging to deal with unseen objects. To tackle this problem, we present a synthetic data generation pipeline that randomizes textures via domain randomization to focus on the shape information. In addition, we propose an RGB-D Fusion Mask R-CNN with a confidence map estimator, which exploits reliable depth information in multiple feature levels. We transferred the trained model to real-world scenarios and evaluated its performance by making comparisons with baselines and ablation studies. We demonstrate that our methods, which use only synthetic data, could be effective solutions for unseen industrial components segmentation.

up
0 users have voted:

Paper Details

Authors:
Jongwon Kim, Raeyoung Kang, Seungjun Choi, Kyoobin Lee
Submitted On:
15 November 2020 - 3:24am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Seunghyeok Back
Paper Code:
3032
Document Year:
2020
Cite

Document Files

PresentationSlides-ICIP2020-Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data.pdf

(7)

Subscribe

[1] Jongwon Kim, Raeyoung Kang, Seungjun Choi, Kyoobin Lee, "Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5554. Accessed: Nov. 29, 2020.
@article{5554-20,
url = {http://sigport.org/5554},
author = {Jongwon Kim; Raeyoung Kang; Seungjun Choi; Kyoobin Lee },
publisher = {IEEE SigPort},
title = {Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data},
year = {2020} }
TY - EJOUR
T1 - Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data
AU - Jongwon Kim; Raeyoung Kang; Seungjun Choi; Kyoobin Lee
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5554
ER -
Jongwon Kim, Raeyoung Kang, Seungjun Choi, Kyoobin Lee. (2020). Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data. IEEE SigPort. http://sigport.org/5554
Jongwon Kim, Raeyoung Kang, Seungjun Choi, Kyoobin Lee, 2020. Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data. Available at: http://sigport.org/5554.
Jongwon Kim, Raeyoung Kang, Seungjun Choi, Kyoobin Lee. (2020). "Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data." Web.
1. Jongwon Kim, Raeyoung Kang, Seungjun Choi, Kyoobin Lee. Segmenting Unseen Industrial Components In A Heavy Clutter Using RGB-D Fusion And Synthetic Data [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5554