SLAM-based 3D surveying
In the last years, we have witnessed a more and more fusion of photogrammetric, computer vision and robotics algorithms (e.g. visual odometry, SLAM, etc.) and their embedding in 3D surveying systems. This allows real-time on-site processing operations with the generation of high-quality 3D results, either using pure image-based solutions or LiDAR-based approaches. We focused on image-based mobile mapping concept, based on visual SLAM solution, to provide innovative and unique features to support the image acquisition and optimising the processing steps.
Our activities include:
Development of a low-cost, lightweight, portable and modular prototype system, called GuPho (Guided Photogrammetry system): it is based on stereo vision, runs a customized vSLAM approach and provide real-time guidance and feedback to the surveyor during the image capturing phase about exposure control, motion blur and ground sample distance. This ensure a more reliable and effective photogrammetric data acquisition and processing.
Testing and validation of the GuPho system in multiple environments, particularly GNSS-denied spaces, like dense forests, underground heritage, military fortifications, tunnels, etc.
Map storage and reuse in a successive survey: this feature allows an user to relocalise the system and map the environment for rephotographing and monitoring specific areas (e.g. a vault with a crack).
Adaptation of GuPho for underwater capabilities: Frog has a spherical dome and similar features of GuPho and can be used both above and under the water with the same calibration when relaxed accuracy requirements exist.
Inclusion of AI methods for semantic image classification, monocular depth estimation, etc.
Inclusion of deep learning features into V-SLAM methods
Representative Publications:
- Menna, F., Torresani, A., Battisti, R., Nocerino, E., Remondino, F., 2022: A modular and low-cost portable VSLAM system for real-time 3D mapping: from indoor and outdoor spaces to underwater environments. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVIII-2/W1-2022, 153–162
- Di Stefano, F., Torresani, A., Farella, E.M., Pierdicca, R., Menna, F., Remondino, F., 2021. 3D Surveying of Underground Built Heritage: Opportunities and Challenges of Mobile Technologies. Sustainability, Vol.13, 13289
- Torresani, A., Menna, F., Battisti, R., Remondino, F., 2021. A V-SLAM Guided and Portable System for Photogrammetric Applications. Remote Sensing, Vol.13(12), 2351
- A. Torresani, S. Rigon, E. M. Farella, F. Menna, F. Remondino, 2021: Unveiling large-scale historical contents with V-SLAM and markerless mobile AR solutions. ISPRS Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-M-1-2021, 761–768
- F. Menna, E. Nocerino, M. M. Nawaf, J. Seinturier, A. Torresani, P. Drap, F. Remondino, B. Chemisky, 2019: Towards real-time underwater photogrammetry for subsea metrology applications. IEEE Oceans 2019 Conference, June 2019, Marseille (France), DOI: 10.1109/OCEANSE.2019.8867285
- A. Torresani and F. Remondino, 2019: Videogrammetry vs photogrammetry for heritage 3D reconstruction. ISPRS Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., Vol. XLII-2/W15, 1135-1140
Funding: internal, EIT-RM