CAM-SegNet: A Context-Aware Dense Material Segmentation Network for Sparsely Labelled Datasets

Yuwen Heng, Yihong Wu, Srinandan Dasmahapatra, Hansung Kim

2022

Abstract

Contextual information reduces the uncertainty in the dense material segmentation task to improve segmentation quality. Typical contextual information includes object, place labels or extracted feature maps by a neural network. Existing methods typically adopt a pre-trained network to generate contextual feature maps without fine-tuning since dedicated material datasets do not contain contextual labels. As a consequence, these contextual features may not improve the material segmentation performance. In consideration of this problem, this paper proposes a hybrid network architecture, the CAM-SegNet, to learn from contextual and material features during training jointly without extra contextual labels. The utility of our CAM-SegNet is demonstrated by guiding the network to learn boundary-related contextual features with the help of a self-training approach. Experiments show that CAM-SegNet can recognise materials that have similar appearances, achieving an improvement of 3-20% on accuracy and 6-28% on Mean IoU.

Paper Citation

in Harvard Style

Heng Y., Wu Y., Dasmahapatra S. and Kim H. (2022). CAM-SegNet: A Context-Aware Dense Material Segmentation Network for Sparsely Labelled Datasets. In Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP; ISBN 978-989-758-555-5, SciTePress, pages 190-201. DOI: 10.5220/0010853200003124

in Bibtex Style

@conference{visapp22,
author={Yuwen Heng and Yihong Wu and Srinandan Dasmahapatra and Hansung Kim},
title={CAM-SegNet: A Context-Aware Dense Material Segmentation Network for Sparsely Labelled Datasets},
booktitle={Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP},
year={2022},
pages={190-201},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010853200003124},
isbn={978-989-758-555-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP
TI - CAM-SegNet: A Context-Aware Dense Material Segmentation Network for Sparsely Labelled Datasets
SN - 978-989-758-555-5
AU - Heng Y.
AU - Wu Y.
AU - Dasmahapatra S.
AU - Kim H.
PY - 2022
SP - 190
EP - 201
DO - 10.5220/0010853200003124
PB - SciTePress