This paper addresses 3-D reconstruction of a marine floating platform using an unmanned surface vessel  (USV). Lidar and sonar sensors equipped on the USV are used to gather volumetric point cloud data of the floating platform structure both above and below the waterline. These measurements are collected in the vehicle-fixed frame, and thus, an accurate estimation of the vehicle 's pose  (i.e., position and attitude) is essential for satisfactory 3-D reconstruction. Because global positioning system signals are severely deteriorated or unavailable around large steel structures, relative navigation with respect to the planar surfaces of their hull structures is performed in the framework of simultaneous localization and mapping. To verify and demonstrate the feasibility of the proposed approach, a field experiment around a full-scale semisubmersible platform was conducted using a custom-designed USV.