BS3D : Building-Scale 3D Reconstruction from RGB-D Images

Janne Mustaniemi*, Juho Kannala, Esa Rahtu, Li Liu, Janne Heikkilä

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review


Various datasets have been proposed for simultaneous localization and mapping (SLAM) and related problems. Existing datasets often include small environments, have incomplete ground truth, or lack important sensor data, such as depth and infrared images. We propose an easy-to-use framework for acquiring building-scale 3D reconstruction using a consumer depth camera. Unlike complex and expensive acquisition setups, our system enables crowd-sourcing, which can greatly benefit data-hungry algorithms. Compared to similar systems, we utilize raw depth maps for odometry computation and loop closure refinement which results in better reconstructions. We acquire a building-scale 3D dataset (BS3D) and demonstrate its value by training an improved monocular depth estimation model. As a unique experiment, we benchmark visual-inertial odometry methods using both color and active infrared images.

Original languageEnglish
Title of host publicationImage Analysis - 23rd Scandinavian Conference, SCIA 2023, Proceedings
EditorsRikke Gade, Michael Felsberg, Joni-Kristian Kämäräinen
Number of pages15
ISBN (Print)978-3-031-31437-7
Publication statusPublished - 2023
MoE publication typeA4 Conference publication
EventScandinavian Conference on Image Analysis - Levi, Kittilä, Finland
Duration: 18 Apr 202321 Apr 2023
Conference number: 23

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13886 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferenceScandinavian Conference on Image Analysis
Abbreviated titleSCIA


  • Depth camera
  • Large-scale
  • SLAM


Dive into the research topics of 'BS3D : Building-Scale 3D Reconstruction from RGB-D Images'. Together they form a unique fingerprint.

Cite this