| Acronym: BI/008/2024 |
| Main Objective: This project aims to investigate the behavior and accuracy of state-of-the-art Simultaneous Localization and Mapping (SLAM) algorithms in challenging underwater cave environments. The research will follow a structured plan comprising four main tasks: (1) the collection, setup, and adaptation of representative SLAM algorithms and 3D reconstruction methods; (2) the curation of a novel underwater cave dataset, under development within the 3D-CAVE FCT project, incorporating visual and inertial data with tailored video sequences to evaluate reconstruction, path estimation, and loop closure accuracy, while also establishing mechanisms for comparison with available ground truths and providing feedback to field teams; (3) a comprehensive experimental study of the curated dataset using the selected algorithms; and (4) dissemination of the findings through the preparation of a scientific article introducing the new underwater cave dataset and its benchmarking potential. |
| Reference: UID/50008/2023 IT. |
| Funding: FCT |
| Approval Date: 12-12-2024 |
| Start Date: 15-12-2024 |
| End Date: 31-12-2025 |
| Team: Hamid Moghadaspour |
| Groups: Multimedia Signal Processing – Lr |
| Partners: Instituto de Telecomunicações |
| Local Coordinator: Hamid Moghadaspour |
|
|
