Enhancing Geometric Edge Details in MVS Reconstruction

Abstract

Abstract. Mesh models generated by multi view stereo (MVS) algorithms often fail to represent in an adequate manner the sharp, natural edge details of the scene. The harsh depth discontinuities of edge regions are eventually a challenging task for dense reconstruction, while vertex displacement during mesh refinement frequently leads to smoothed edges that do not coincide with the fine details of the scene. Meanwhile, 3D edges have been used for scene representation, particularly man-made built environments, which are dominated by regular planar and linear structures. Indeed, 3D edge detection and matching are commonly exploited either to constrain camera pose estimation, or to generate an abstract representation of the most salient parts of the scene, and even to support mesh reconstruction. In this work, we attempt to jointly use 3D edge extraction and MVS mesh generation to promote edge detail preservation in the final result. Salient 3D edges of the scene are reconstructed with state-of-the-art algorithms and integrated in the dense point cloud to be further used in order to support the mesh triangulation step. Experimental results on benchmark dataset sequences using metric and appearance-based measures are performed in order to evaluate our hypothesis.

Vision through Obstacles—3D Geometric Reconstruction and Evaluation of Neural Radiance Fields (NeRFs)

“…Based on the projection between a set of overlapping images taken from different positions, the depth values of each pixel are calculated, which is time-consuming and restricts the completeness of the reconstructed point cloud [9]. Since MVS often fails in estimating correct depth pixel values in low-textured areas of the scene, applying 3D edge extraction can support the mesh triangulation step [10] and preserve geometric details. However, the quality of the reconstructed mesh strongly depends on the accuracy of the extracted 3D edges.…”

Geometric Accuracy Analysis Between Neural Radiance Fields (Nerfs) and Terrestrial Laser Scanning (Tls)

“…Generating a dense point cloud of a scene based on camera parameters and poses in 3D space is initially addressed by Multi-View Stereo (MVS) algorithms. MVS allows high reconstruction accuracy when the surfaces are textured, thus the main challenge lies in computing precise corresponding points between images of the scene (Stathopoulou et al, 2021).…”