Multi-View Stereo Revisited

Abstract

We present an extremely simple yet robust multi-view stereo algorithm and analyze its properties. The algorithm first computes individual depth maps using a window-based voting approach that returns only good matches. The depth maps are then merged into a single mesh using a straightforward volumetric approach. We show results for several datasets, showing accuracy comparable to the best of the current state of the art techniques and rivaling more complex algorithms.

Video-based, real-time multi-view stereo

“…We start by referring to the MVS evaluation by [23]. Looking at the top performers in that evaluation, we can distinguish two main trends: region growing methods [8,11,12,20] and occlusion-robust photo-consistency methods [3,5,10,13,18,27]. The best performing region growing method [8] uses a combination of photo-consistency based patch fitting, growing and filtering in order to reconstruct the scene of interest.…”

Large Scale Scenes Reconstruction from Multiple Views

“…outdoor architectural scenes) are those representing geometry by several depth maps [16,17,18]. However, their performance for complete reconstruction seems to be lower than previously discussed approaches, either in terms of accuracy or the completeness of the obtained model.…”

Generic Scene Recovery Using Multiple Images

“…It is split into two parts: the pixels corresponding to the foreground, π iF = π i ∩ Π i (S), and the other points π iB = π i \ π iF . I i : π i → R c is the image of the true scene, captured by the i th camera 3 . I is the set of input images and I iF and I iB are the restrictions of the function I i to π iF and π iB , respectively.…”

“…On the other hand, for a long time, the estimation of surface radiance/reflectance was secondary. Even some recent works [2][3][4][5] compute the 3D shape without considering radiance estimation. However, radiance/reflectance estimation has become a matter of concern in multiview reconstruction scenarios in the last decade [6][7][8].…”