A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Semi-supervised domain adaptation for segmentation models on different monitoring settings
Abstract The performance of deep learning models could easily degrade even with slight changes in monitoring settings and environments. Although previous studies have addressed such problems with domain adaptation (DA) methods, this study found that even the state-of-the-art DA methods could not achieve decent adaptation performance in the construction domain. To address the problem, this study presents a novel semi-supervised DA method for semantic segmentation that is built on data augmentation, an unsupervised DA method, and knowledge distillation. Experiments were conducted on the dataset from one construction site (the source domain) and three different construction sites (the target domains). The experimental results demonstrated the effectiveness of the proposed method: the performance is comparable to a model trained with the data in the target domain. The major finding is that DA can be significantly improved when unsupervised DA methods are used with the target domain-specific data augmentation for construction site scenes.
Graphical abstract Display Omitted
Highlights A semi-supervised domain adaptation method was proposed for construction site scenes. The segmentation performance was improved in the target domain without annotations. The performance improved by 12.80%, 21.39%, and 32.08% for the target domains. Demonstrated the potential of DA to reduce training data in new scenes. A dataset is presented for future study to advance DA methods in jobsite monitoring.
Semi-supervised domain adaptation for segmentation models on different monitoring settings
Abstract The performance of deep learning models could easily degrade even with slight changes in monitoring settings and environments. Although previous studies have addressed such problems with domain adaptation (DA) methods, this study found that even the state-of-the-art DA methods could not achieve decent adaptation performance in the construction domain. To address the problem, this study presents a novel semi-supervised DA method for semantic segmentation that is built on data augmentation, an unsupervised DA method, and knowledge distillation. Experiments were conducted on the dataset from one construction site (the source domain) and three different construction sites (the target domains). The experimental results demonstrated the effectiveness of the proposed method: the performance is comparable to a model trained with the data in the target domain. The major finding is that DA can be significantly improved when unsupervised DA methods are used with the target domain-specific data augmentation for construction site scenes.
Graphical abstract Display Omitted
Highlights A semi-supervised domain adaptation method was proposed for construction site scenes. The segmentation performance was improved in the target domain without annotations. The performance improved by 12.80%, 21.39%, and 32.08% for the target domains. Demonstrated the potential of DA to reduce training data in new scenes. A dataset is presented for future study to advance DA methods in jobsite monitoring.
Semi-supervised domain adaptation for segmentation models on different monitoring settings
Hong, Yeji (author) / Chern, Wei-Chih (author) / Nguyen, Tam V. (author) / Cai, Hubo (author) / Kim, Hongjo (author)
2023-01-27
Article (Journal)
Electronic Resource
English
Instance segmentation using semi-supervised learning for fire recognition
| Elsevier | 2022