Spatiotemporal correlation using inverse triangulation technique with a variable pseudorandom pattern projector for 3D stereo measurement

In this paper, we propose an innovative 3D reconstruction approach based on stereo vision that combines inverse triangulation with spatiotemporal correlation algorithms. The inverse triangulation technique allows the correspondence search to be performed in an ordered manner within object space. In this way, the proposed approach performs the correspondence search using well-defined three-dimensional regions in object space. These regions are structured as regular grids centered on each analyzed point within the measurement volume. The spatiotemporal correlation across the projected regions determines the optimal value, resulting in a structured and reliable 3D reconstruction. This proposed approach is intended for future applications focused on the underwater inspection of offshore structures in the oil and gas industry.