| Acronym: 3D-CAVE |
| Main Objective: The main objective of this work is the study and analysis of the performance of Visual-Inertial Simultaneous Localization and Mapping (VI-SLAM) methods in underwater environments, with the long-term goal of enabling the development of a complete 3D mapping ecosystem for underwater caves, currently under construction within the scope of the 3D-CAVE project. The proposed work plan encompasses several activities: (i) the analysis and evaluation of state-of-the-art VI-SLAM algorithms on real underwater cave video sequences acquired by the 3D-CAVE project team; (ii) the proposal and implementation of new modules to enhance the performance of VI-SLAM algorithms in underwater scenarios; (iii) the identification of the most computationally demanding algorithmic components with the aim of accelerating them through dedicated hardware; (iv) the design and development of hardware modules for VI-SLAM acceleration, targeting their deployment across different ranges of FPGA devices; and (v) the dissemination of the main results in leading scientific venues in the field. |
| Reference: 2022.04020.PTDC |
| Funding: FCT |
| Approval Date: 10-10-2023 |
| Start Date: 18-10-2023 |
| End Date: 18-10-2024 |
| Team: Hamid Moghadaspour |
| Groups: Multimedia Signal Processing – Lr |
| Partners: INESC-ID |
| Local Coordinator: Hamid Moghadaspour |
|
|
