About this Research Topic
Humans spend most of their time indoors (i.e. in office buildings, shopping malls, big transport hubs, etc.). Recent years have witnessed rapid advancement of indoor location-based services (LBS) with the continuous evolution of mobile devices and communication technologies. Indoor LBS has many prospective applications, such as in smart cities, intelligent transportation, emergency management, and logistics management. According to IMARC, the indoor LBS market value is estimated to reach $34 Billion by 2027. Updated, detailed, and semantically rich 2D/3D indoor maps and models are indispensable parts of Indoor LBS applications. However, currently, only a small fraction of the millions of indoor environments have been mapped. Moreover, these existing maps are often out-of-date, and consequently, they do not represent the as-is condition of the environment.
In the last few years, there has been intense research activity towards the automatic reconstruction of indoor environments. Many novel solutions have been proposed, e.g., parsing floor plan images, leveraging special sensors (e.g., laser, camera, depth camera, and sonar) equipped on a robot or backpack platform, and crowdsourcing (e.g., to collect volunteers’ trajectories estimated based on their smartphones). However, there are still specific challenges in reconstructing the geometry of indoor environments. For example, the aforementioned solutions suffer from the varied representation styles and inaccessibility of the floor plans, the complicated indoor structure layouts, clutter, and occlusions, and the quality issue of crowdsourced data, respectively. Moreover, the issue of automatically assigning semantic labels to indoor objects (e.g., room, shop, and furniture) still has not been completely solved. Furthermore, unified storage and exchange possibilities of reconstructed 3D models, through the use of known BIM and GIS open standards (e.g. IFC, CityGML, IndoorGML, LADM) can be improved. Therefore, this Research Topic provides a timely opportunity to discuss and summarize the latest developments in this area.
This Research Topic welcomes original research papers and review papers offering new insights into indoor mapping and modeling. Topics of interest include, but are not limited to:
• New sensing technologies for indoor mapping;
• Indoor reconstruction;
• Simultaneous localization and mapping;
• Crowdsourcing-based indoor mapping;
• Floor plan parsing;
• Geometric evaluation of indoor mapping systems;
• Indoor positioning;
• Neural inertial localization;
• Indoor data structures and models;
• BIM feature extraction on indoor 3D point clouds;
• Indoor landmark detection;
• Semantic labeling of indoor environment;
• Semantic parsing.
In the last few years, there has been intense research activity towards the automatic reconstruction of indoor environments. Many novel solutions have been proposed, e.g., parsing floor plan images, leveraging special sensors (e.g., laser, camera, depth camera, and sonar) equipped on a robot or backpack platform, and crowdsourcing (e.g., to collect volunteers’ trajectories estimated based on their smartphones). However, there are still specific challenges in reconstructing the geometry of indoor environments. For example, the aforementioned solutions suffer from the varied representation styles and inaccessibility of the floor plans, the complicated indoor structure layouts, clutter, and occlusions, and the quality issue of crowdsourced data, respectively. Moreover, the issue of automatically assigning semantic labels to indoor objects (e.g., room, shop, and furniture) still has not been completely solved. Furthermore, unified storage and exchange possibilities of reconstructed 3D models, through the use of known BIM and GIS open standards (e.g. IFC, CityGML, IndoorGML, LADM) can be improved. Therefore, this Research Topic provides a timely opportunity to discuss and summarize the latest developments in this area.
This Research Topic welcomes original research papers and review papers offering new insights into indoor mapping and modeling. Topics of interest include, but are not limited to:
• New sensing technologies for indoor mapping;
• Indoor reconstruction;
• Simultaneous localization and mapping;
• Crowdsourcing-based indoor mapping;
• Floor plan parsing;
• Geometric evaluation of indoor mapping systems;
• Indoor positioning;
• Neural inertial localization;
• Indoor data structures and models;
• BIM feature extraction on indoor 3D point clouds;
• Indoor landmark detection;
• Semantic labeling of indoor environment;
• Semantic parsing.
Keywords: Indoor mapping, Indoor modeling, Building reconstruction, Indoor LBS, Semantic labeling, Indoor positioning, BIM
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
