3d Reconstruction of Unstable Underwater Environment With SFM Using Slam

Abstract

Abstract. The underwater environment has substantial properties for underwater research such as marine archaeology, monitoring coral reefs, and shipwrecks. SfM, as a major step of photogrammetry, has been widely used in the field. For a high 3D construction quality, images must have a clear visual sight environment and known orientations of the images. However, underwater images have various types of visual disturbances, but also GPS/INS, commonly used on the ground, are not accepted. Finding more feature points or using more images for SfM are solutions to the problems. However, these methods take high computational costs. An alternative to this problem is to provide the known orientations of the images. For a solution to provide known orientations of images, the presented method in this study uses visual SLAM that processes the localization of a vehicle system and mapping of surroundings. The experiment aims to verify whether SLAM improves the quality of underwater 3D reconstruction and the computation efficiency of SfM. We examine the two aqualoc datasets with the results of the number of cloud points, SfM processing time, and the number of matched images/total images and mean reprojection errors. The outcome shows SLAM-determined orientations improved the quality of 3D reconstruction and the computation efficiency of SfM with results of the increased number of point clouds and the decreased processing time.

Overview of Underwater 3D Reconstruction Technology Based on Optical Images

“…Furthermore, since most techniques are performed under the assumption of planarity of the refraction interface, they proposed a technique to relax this assumption using soft constraints in order to apply this technique to natural water surfaces. Jeon and Lee [102] proposed the use of visual simultaneous localization and mapping (SLAM) to handle the localization of vehicle systems and the mapping of the surrounding environment. The orientation determined using SLAM improves the quality of 3D reconstruction and the computational efficiency of SfM, while increasing the number of point clouds and reducing the processing time.…”

Efficient Structure from Motion for Large-Size Videos from an Open Outdoor UAV Dataset

“…To process large-size videos in real-time, SLAM systems estimate camera poses and build maps on downsampled images, which is more efficient but results in lower localization accuracy than SfM methods. Some researchers [17,18] have attempted to utilize VSLAM to assist the SfM method for reconstruction. However, these methods only utilize estimated camera poses from a SLAM system.…”