StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM Modeling

The integration of computer vision and deep learning into Building Information Modeling (BIM) workflows has created a growing need for structured datasets that enable the semantic segmentation of indoor building elements. This paper presents StructScan3D v1, the first version of an RGB-D dataset spe...

Full description

Saved in:
Bibliographic Details
Main Authors: Ishraq Rached, Rafika Hajji, Tania Landes, Rashid Haffadi
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/11/3461
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849721842893324288
author Ishraq Rached
Rafika Hajji
Tania Landes
Rashid Haffadi
author_facet Ishraq Rached
Rafika Hajji
Tania Landes
Rashid Haffadi
author_sort Ishraq Rached
collection DOAJ
description The integration of computer vision and deep learning into Building Information Modeling (BIM) workflows has created a growing need for structured datasets that enable the semantic segmentation of indoor building elements. This paper presents StructScan3D v1, the first version of an RGB-D dataset specifically designed to facilitate the automated segmentation and modeling of architectural and structural components. Captured using the Kinect Azure sensor, StructScan3D v1 comprises 2594 annotated frames from diverse indoor environments, including residential and office spaces. The dataset focuses on six key building elements: walls, floors, ceilings, windows, doors, and miscellaneous objects. To establish a benchmark for indoor RGB-D semantic segmentation, we evaluate D-Former, a transformer-based model that leverages self-attention mechanisms for enhanced spatial understanding. Additionally, we compare its performance against state-of-the-art models such as Gemini and TokenFusion, providing a comprehensive analysis of segmentation accuracy. Experimental results show that D-Former achieves a mean Intersection over Union (mIoU) of 67.5%, demonstrating strong segmentation capabilities despite challenges like occlusions and depth variations. As an evolving dataset, StructScan3D v1 lays the foundation for future expansions, including increased scene diversity and refined annotations. By bridging the gap between deep learning-driven segmentation and real-world BIM applications, this dataset provides researchers and practitioners with a valuable resource for advancing indoor scene reconstruction, robotics, and augmented reality.
format Article
id doaj-art-dd1ba4f18ea64c00b27f6f6f79371ec6
institution DOAJ
issn 1424-8220
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-dd1ba4f18ea64c00b27f6f6f79371ec62025-08-20T03:11:32ZengMDPI AGSensors1424-82202025-05-012511346110.3390/s25113461StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM ModelingIshraq Rached0Rafika Hajji1Tania Landes2Rashid Haffadi3College of Geomatic Sciences and Surveying Engineering, Institute of Agronomy and Veterinary Medicine, Rabat 6202, MoroccoCollege of Geomatic Sciences and Surveying Engineering, Institute of Agronomy and Veterinary Medicine, Rabat 6202, MoroccoICube Laboratory UMR 7357, Photogrammetry and Geomatics Group, National Institute of Applied Sciences (INSA Strasbourg), 24, Boulevard de la Victoire, 67084 Strasbourg, FranceGEOPTIMA, B4, Med El Amraoui Street, Corner of Sebou Street, Office 4, Kenitra, MoroccoThe integration of computer vision and deep learning into Building Information Modeling (BIM) workflows has created a growing need for structured datasets that enable the semantic segmentation of indoor building elements. This paper presents StructScan3D v1, the first version of an RGB-D dataset specifically designed to facilitate the automated segmentation and modeling of architectural and structural components. Captured using the Kinect Azure sensor, StructScan3D v1 comprises 2594 annotated frames from diverse indoor environments, including residential and office spaces. The dataset focuses on six key building elements: walls, floors, ceilings, windows, doors, and miscellaneous objects. To establish a benchmark for indoor RGB-D semantic segmentation, we evaluate D-Former, a transformer-based model that leverages self-attention mechanisms for enhanced spatial understanding. Additionally, we compare its performance against state-of-the-art models such as Gemini and TokenFusion, providing a comprehensive analysis of segmentation accuracy. Experimental results show that D-Former achieves a mean Intersection over Union (mIoU) of 67.5%, demonstrating strong segmentation capabilities despite challenges like occlusions and depth variations. As an evolving dataset, StructScan3D v1 lays the foundation for future expansions, including increased scene diversity and refined annotations. By bridging the gap between deep learning-driven segmentation and real-world BIM applications, this dataset provides researchers and practitioners with a valuable resource for advancing indoor scene reconstruction, robotics, and augmented reality.https://www.mdpi.com/1424-8220/25/11/3461semantic segmentationRGB-DKinect AzureBIMstructural elementsdataset
spellingShingle Ishraq Rached
Rafika Hajji
Tania Landes
Rashid Haffadi
StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM Modeling
Sensors
semantic segmentation
RGB-D
Kinect Azure
BIM
structural elements
dataset
title StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM Modeling
title_full StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM Modeling
title_fullStr StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM Modeling
title_full_unstemmed StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM Modeling
title_short StructScan3D v1: A First RGB-D Dataset for Indoor Building Elements Segmentation and BIM Modeling
title_sort structscan3d v1 a first rgb d dataset for indoor building elements segmentation and bim modeling
topic semantic segmentation
RGB-D
Kinect Azure
BIM
structural elements
dataset
url https://www.mdpi.com/1424-8220/25/11/3461
work_keys_str_mv AT ishraqrached structscan3dv1afirstrgbddatasetforindoorbuildingelementssegmentationandbimmodeling
AT rafikahajji structscan3dv1afirstrgbddatasetforindoorbuildingelementssegmentationandbimmodeling
AT tanialandes structscan3dv1afirstrgbddatasetforindoorbuildingelementssegmentationandbimmodeling
AT rashidhaffadi structscan3dv1afirstrgbddatasetforindoorbuildingelementssegmentationandbimmodeling