Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs
This paper presents a new framework to classify floor plan elements and represent them in a vector format. Unlike existing approaches using image-based learning frameworks as the first step to segment the image pixels, we first convert the input floor plan image into vector data and utilize a graph...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/10/2/97 |
| _version_ | 1797395747846488064 |
|---|---|
| author | Jaeyoung Song Kiyun Yu |
| author_facet | Jaeyoung Song Kiyun Yu |
| author_sort | Jaeyoung Song |
| collection | DOAJ |
| description | This paper presents a new framework to classify floor plan elements and represent them in a vector format. Unlike existing approaches using image-based learning frameworks as the first step to segment the image pixels, we first convert the input floor plan image into vector data and utilize a graph neural network. Our framework consists of three steps. (1) image pre-processing and vectorization of the floor plan image; (2) region adjacency graph conversion; and (3) the graph neural network on converted floor plan graphs. Our approach is able to capture different types of indoor elements including basic elements, such as walls, doors, and symbols, as well as spatial elements, such as rooms and corridors. In addition, the proposed method can also detect element shapes. Experimental results show that our framework can classify indoor elements with an F1 score of 95%, with scale and rotation invariance. Furthermore, we propose a new graph neural network model that takes the distance between nodes into account, which is a valuable feature of spatial network data. |
| first_indexed | 2024-03-09T00:39:01Z |
| format | Article |
| id | doaj.art-f6e9cc5e7e874f82b763664fa44a36f1 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-09T00:39:01Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-f6e9cc5e7e874f82b763664fa44a36f12023-12-11T17:58:08ZengMDPI AGISPRS International Journal of Geo-Information2220-99642021-02-011029710.3390/ijgi10020097Framework for Indoor Elements Classification via Inductive Learning on Floor Plan GraphsJaeyoung Song0Kiyun Yu1Department of Civil and Environmental Engineering, Seoul National University, Seoul 08826, KoreaDepartment of Civil and Environmental Engineering, Seoul National University, Seoul 08826, KoreaThis paper presents a new framework to classify floor plan elements and represent them in a vector format. Unlike existing approaches using image-based learning frameworks as the first step to segment the image pixels, we first convert the input floor plan image into vector data and utilize a graph neural network. Our framework consists of three steps. (1) image pre-processing and vectorization of the floor plan image; (2) region adjacency graph conversion; and (3) the graph neural network on converted floor plan graphs. Our approach is able to capture different types of indoor elements including basic elements, such as walls, doors, and symbols, as well as spatial elements, such as rooms and corridors. In addition, the proposed method can also detect element shapes. Experimental results show that our framework can classify indoor elements with an F1 score of 95%, with scale and rotation invariance. Furthermore, we propose a new graph neural network model that takes the distance between nodes into account, which is a valuable feature of spatial network data.https://www.mdpi.com/2220-9964/10/2/97floor plan analysisvectorizationgraph neural networkindoor spatial data |
| spellingShingle | Jaeyoung Song Kiyun Yu Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs ISPRS International Journal of Geo-Information floor plan analysis vectorization graph neural network indoor spatial data |
| title | Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs |
| title_full | Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs |
| title_fullStr | Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs |
| title_full_unstemmed | Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs |
| title_short | Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs |
| title_sort | framework for indoor elements classification via inductive learning on floor plan graphs |
| topic | floor plan analysis vectorization graph neural network indoor spatial data |
| url | https://www.mdpi.com/2220-9964/10/2/97 |
| work_keys_str_mv | AT jaeyoungsong frameworkforindoorelementsclassificationviainductivelearningonfloorplangraphs AT kiyunyu frameworkforindoorelementsclassificationviainductivelearningonfloorplangraphs |